X / CO ^ o ™ 2 avaan^LIBRARI esotsmithsonian_ institution ^NounxixsNiXviNOSHXinis^sa 1 ava ^ ., w /S3££&\ “ /£§®D\ « w'v - ITUTION ^NOIXfUIXSNI ^ NVINOSHXIIAIs'0 S3 I ava BIUIB RAR I ES SMITHSONIAN INSTITUT CO . 2 X. ^ , •~p> CO w w V w /&2gfof. m zT?°»ox w °A h m, t: ~i /&&Wep\ — WW vj\ o * ..VSr «• o avd a ll"”u B RAR I ES SMITHSONIAN^INSTITUTION NOlinillSNI^NVINOSHimS^Sa lava o H 3 - ww “ w ^ -- vo; TITUTION^ NOliniliSNI- NVINOSHIIWS^SB I BVB Q H L I B RAR I ES^SMITHSONIAN INSTITU' JS\. s ,< S ,< a&JVo E E m'vc i ava a n libraries smithsonian institution NouniiiSNi nvinoshiiiais saiava — CO ^ ^5 X CO w 4 co 'tA — /5 •« . /‘S(PfkK/\ ™ J '’V\S"N&8\ o x^yv pc^ — o _ ^ TITUTIOn"’ N0llfUllSNl"JNV!N0SHilWS2S3 I a Va a II LI B R AR I ES SMITHSONIANJNSTITU' z r- , z r- z m Vtpr w - N^Sp?' g ' rn X«.i55«j I ava 0 IT- LI B R AR I ESWSMITHSONIAN”lNSTITUTIONWNOIinilXSNI-NVINOSHXIWS S3 I UV0 z ^ z . w z ^ < s x5®r!5x < A. S S * 5 !ik'W S \^»U.|W?/ 5 %W i WW" '§§W‘ - * ' ^ i Cwp i i, . , _ 2 • ' N^S»' > XjVASt^X ^ Vteyos^ TITUTION^NOIXnXIXSNlf NVIN0SHXIWSWS3 I a Va a nf LI B RAR I ES^SMITHSONIANJNSTITU UJ 5 'X. U1 m y^5s5oX w ^7/^‘y S GO 5 RAR I ESWSMITHSONIANZ INSTITUTION NOIlfUllSNI NVINOSHXIWS^SB I ava 3 ll^ LI B RAF CO 5 M ~ ,.v . co w x ,, <3 /5$§2$i\ w /^S^5k F3 “* xs^^r 2- '•'/$«£$ 3 (£bz> i*\ “ ($tr ’'z&x — ; «* 2 (ie 3^ < sfe 3° ^ «8§k^ < i°i o: 'WMm S WfeuiT?/ K WmM S « W ® 5 2 X^j^y/ 5 2 xasiSix o iniliSN|JNVINOSHimS2S3iavaan"JLIBRARIESZSMITHSONIAN“lNSTITUTION NOlifUl I 1 1 0%) I #^.i ^ ■sTrrz-tfj "J pi N^vosw^y “ * ^ ^ rn 'iftos«s2' "J xjwsiv I RAR I ES^SMITHSONIAn” INSTITUTION^NOIXflillSNI- NVIN0SH1IWS S3 I UVH 8 11 ~Lf B RAF w 55?iroX < * “ § g ip. < ’■* 1 I O 3 i||fe|fe o X MP^fe- o o 1 £ -^Si 8 Mw$- x (t§. Jl) 8 > s '" V?-'' > ’ 2 X^QSKjy > s IDillSNI NVIN0SHXUNSC°S3 lava BIT^LI B RAR I ES^SMITHSONIAN INSTITUTION ^NOIXfUl GO ~ (O CD CD o\ ~ d) H Z «j z m _ ___ 3 RAR I ES SMITHSONIAN INSTITUTION NOIXFUIXSNI NVINOSHXIWS SSiavaaiT LIB RAF z <~ z ~ rtv«OA>v O . Z tyX O — Ay ... rS ™ GO ^ IXnXIXSNI- NVINOSHXIINS S3 I BVB a n~LI B RAR I ES^SMITHSONIAN “INSTITUTION N0IJ-ni Z CO Z *,♦♦.. CO jr CO < S *< < jjEN\, S t\ -r! I I I I XXU&& t 5 ^ H z w V Z V> BRARIES SMITHSONIAN INSTITUTION NOIXIUIXSNI NVINOSHXIWS S3iavaaiT_ CO —* CO ZZ CO LIBRAI O N<£>xixy>^ q O ixniusNi^NviNOSHiiws^sa lavaan^LiBRARi esz Smithsonian-1! nstitution noixoxi r* „ z r- Z r- ^ CD > ?o K'asvv^x rn ^ xjvasv^x rn B RAR 1 ES SMITHS0NIAN_INSTITUTI0NWN0IXnXIXSNI-NVIN0SHXIINSS3 I ava a IT ,LI B RAF ^ ^ . z « u> z w r&m X CO o ^ z 4 / ' ' t \mmKp/ E > s *'■' \^v >■' 25 >’ lifUUSNI_NVIN0SHimS^S3 I H VH 8 ll^LI B RAR I ES^SMITHSONIAN^INSTITUTION NO*10-11 ^nr^A h zi - The Journal of ARACHNOLOGY OFFICIAL ORGAN OF THE AMERICAN ARACHNOLOGICAL SOCIETY VOLUME 9 WINTER 1981 NUMBER 1 THE JOURNAL OF ARACHNOLOGY EDITOR: Oscar F. Francke, Texas Tech University. ASSOCIATE EDITOR’. B. J. Kaston, San Diego State University. EDITORIAL ASSISTANT: James C. Cokendolpher, Texas Tech University. EDITORIAL BOARD: Charles D. Dondale, Agriculture Canada. William G. Eberhard, Universidad de Costa Rica. Maria E. Galiano, Museo Argentino de Ciencias Naturales. Willis J. Gertsch, American Museum of Natural History. Neil F. Hadley, Arizona State University. B. J. Kaston, San Diego State University. Herbert W. Levi, Harvard University. Emilio A. Maury, Museo Argentino de Ciencias Naturales. William B. Muchmore, University of Rochester. Martin H. Muma, Western New Mexico University. William B. Peck, Central Missouri State University. Norman I. Platnick, American Museum of Natural History. Susan E. Riechert, University of Tennessee. Michael E. Robinson, Smithsonian Tropical Research Inst. Jerome S. Rovner, Ohio University. William A. Shear, Hampden-Sydney College. Carlos E. Valerio, Universidad de Costa Rica. Stanley C. Williams, San Francisco State University. THE JOURNAL OF ARACHNOLOGY is published in Winter, Spring, and Fall by The American Arachno logical Society in cooperation with The Graduate School, Texas Tech University. Individual subscriptions, which include membership in the Society, are $20.00 for regular members, $12.00 for student members. Institutional subscriptions to The Journal are $25.00. Correspondence concerning subscription and membership should be addres- sed to the Membership Secretary (see back inside cover). Back issues of The Journal are available from Dr. William B. Peck, Department of Biology, Central Missouri State Univer- sity, Warrensburg, Missouri 64093, U.S.A., at $5.00 for each number. Remittances should be made payable to The American Arachnological Society. Change of address notices must be sent to both the Secretary and the Membership Secretary. Manuscripts for THE JOURNAL OF ARACHNOLOGY are acceptable only from cur- rent members of the Society, and there are no page charges. Manuscripts must be type- written double or triple spaced on 8.5 in. by 11 in. bond paper with ample margins, and may be written in the following languages: English, French, Portuguese, and Spanish. Contributions dealing exclusively with any of the orders of Arachnida, excluding Acari, will be considered for publication. Papers of a comparative nature dealing with chelicer- ates in general, and directly relevant to the Arachnida are also acceptable. Detailed instructions for the preparation of manuscripts appear in the Fall issue of each year, and can also be obtained from the Editor and the Associate Editor. Manuscripts that do not follow those instructions will be returned to the author(s) without the benefit of review. Manuscripts and all related correspondence must be sent to Dr. B. J. Kaston, 5484 Hewlett Drive, San Diego, California 921 15, U.S.A. Cokendolpher, J. C. 1981. Revision of the genus Trachyrhinus Weed (Opiliones, Phalangioidea). J. Arachnol., 9:1-18. REVISION OF THE GENUS TRACHYRHINUS WEED (OPILIONES, PHALANGIOIDEA) James C. Cokendolpher The Museum and Department of Biological Sciences Texas Tech University, Lubbock, Texas 79409 ABSTRACT The western North American phalangioid genus Trachyrhinus is redefined and diagnosed. Trachy- rhinus favosus (Wood) and T. marmoratus Banks are redescribed and illustrated. A neotype for T. favosus and a lectotype for T. marmoratus are designated. Four new species from New Mexico, Texas and northeastern Mexico are described. A key to the species of Trachyrhinus is provided. INTRODUCTION Considerable confusion has surrounded the identity of members of the genus Trachy- rhinus. This in part was due to the difficulty in obtaining type specimens, but to a greater extent to inadequate original descriptions of the two known species: Trachyrhinus favosus (Wood) and Trachyrhinus marmoratus Banks. A third species, Trachyrhinus sono- ranus , was described by Chamberlin (1925), but was synonymized with T. marmoratus by Goodnight and Goodnight (1946). In addition to the inadequate descriptions, the long used character of coxal coloration has proved to be of little value. This paper redescribes known species and describes four new species. METHODS With the growing understanding of the importance of structural measurements, it is necessary to standardize methods used in making measurements. Throughout the descrip- tions the following methods were used. All measurements (means in parentheses) are in mm and were made using a binocular microscope equipped with an ocular micrometer. Body— Total length measured from dorsal view, from anterior tip of supracheliceral lamellae to posterior tip of abdomen; greatest width, not including coxae, measured from dorsal view; maximum height (abdomen) measured from lateral view. Ocular tubercle— All measurements from dorsal view, spine and tubercle lengths not included in measurements. Penis— Length and width measurements as in figure 1 . Palpi— Lengths as in figures 2 and 3. Genital operculum— Length and width measurements as in figure 4. 2 THE JOURNAL OF ARACHNOLOGY Leg segments— Lengths measured along dorsal surface from a lateral view. Abbreviations for collections from which specimens were examined, or deposited in, are listed in the acknowledgments. Specimens in The Museum, Texas Tech University are listed TTU; specimens in my personal collection are JCC. GENUS TRACHYRHINUS WEED Phalangium : Wood 1871:28 (in part); Underwood 1885:168 (in part); Weed 1889:105 (in part). Astrobunus : Weed 1890:914 (in part). Trachyrhinus Weed 1892a:529, 1892b: 193, 1893:287; Banks 1893:206, 1894:145, 1901a:673, 1901b:593, 1901c:588; Scheffer 1906:128; Roewer 1910:259, 1923:872, 1957:356; Chamberlin 1925:172; Goodnight and Goodnight 1942:15, 1946:7; Comstock 1948:67; Milstead 1958:445; Katayama and Post 1974:8. Type species .-Phalangium favosum Wood, by original designation. Comments and diagnosis.— Within the genus Trachyrhinus are apparently two species groups. One group is easily recognized by the presence of a palpal apophysis and by having the alate portion of the penis large and expanded. The second group is characterized by lacking a palpal apophysis, penis with small alate portion, reduced number and size of lateral tubercles on the ocular tubercle, and by having the distal portions of the palpal tarsi slightly expanded. Until the present all described species of A Figs. \ A. -Trachyrhinus: 1, penis, dorsal view ( a = total length, b = width at midshaft); 2, palpus, dorsal view (a = patellar apophysis length); 3, palpus, lateral view (a = femur length, b = patella length, c = tibia length, d = tarsus length); 4, genital operculum, ventral view (a = total length, b = width at neck, c = width at base). COKENDOLPHER-REVISION OF TRACHYRHINUS 3 Trachyrhinus belonged in the group with apophyses, rendering previous generic diagnoses inadequate. The genus Trachyrhinus is diagnosed as follows: Trachyrhinus differs from all other known genera of Phalangioidea by having at least one pseudoarticulary nodule in femora II, tibiae II with pseudosegments, palpal claw smooth, and chelicerae with a tooth ventrally on first segment. Description.— Medium sized phalangioids with hard, coarsely punctate bodies (Figs. 9-11); dorsum with white opalescent spots; with poorly developed lateral sclerites in some specimens. Ocular tubercle approximately equal in length and width, with two well Figs. 5 -11. -Trachyrhinus: 5, femur II, T. maramoratus", 6, femur II, female T. horneri ; 7, seminal receptacle, T. dicropalpus’, 8, ovipositor, T. dicropalpus ; 9-11, male T. marmoratus, 9 dorsal view, 10 lateral view, 1 1 ventral view. 4 THE JOURNAL OF ARACHNOLOGY developed rows of pointed tubercles. Chelicerae not enlarged, with tooth on first segment ventrally. Supracheliceral lamellae in the form of two plates. Scent gland pores small, oval to slightly elongated. Coxae, except posterior portions of III, and genital operculum with lateral rows of tri-pointed denticles; rows on genital operculum rarely reduced or absent. Coxae III and IV enlarged in males (Figs. 10, 11). Legs long, covered with spines and tubercles; femora I equal to or shorter than length of body, II with one or two pseudo- articulary nodules (Figs. 5, 6); tibiae II with pseudosegments. Palpi often with an apophysis on inner distal margins of patellae; tarsi with ventral rows of denticles in males, unarmed in females; claw smooth. Penis alate; shaft long and thin, contracted and bent slightly anterior to alate portion, ending in a sharp tip; alate portion consisting of two sacs opening distally. Ovipositor as in figure 8; seminal receptacle varies slightly between species, general form as in figure 7. Distribution.— Northern Mexico and western United States (Figs. 12-13). Remarks.— The male of Trachyrhinus mesillensis , n. sp. is unknown and will not be listed in the following key. As the female of this species belongs to the group without palpal apophyses the male presumably also has no palpal apophyses. KEY TO ADULT TRACHYRHINUS 1 . Palpal patellae with an apophysis 6 Palpal patellae without an apophysis 2 2. Males 3 Females 4 3. Femora II less than 6.5 mm in length T. horneri Femora II greater than 7.0 mm in length T. rectipalpus 4. Femora II with two pseudoarticulary nodules (Fig. 6), less than 6.5 mm in length . . . T. horneri Femora II with one pseudoarticulary nodule (Fig. 5), greater than 8.0 mm in length. . . 5 5. Femora II less than 10.5 mm in length T. rectipalpus Femora II greater than 1 1 .5 mm in length T. mesillensis 6. Patellar apophysis greater than 0.23 mm in length T. dicropalpus Patellar apophysis less than 0.18 mm in length 7 7. Palpal tarsi with only ventral row of denticles, bare in females (Figs. 23, 25) T. marmoratus Palpal tarsi with many small obtuse tubercles ventrally, reduced in females, as well as ventral rows of denticles in males (Figs. 17, 19) T. favosus Trachyrhinus favosus (Wood) Figs. 12, 14-19 Phalangium favosum Wood 187 1:28; Underwood 1885:168; Weed 1889: 105; Banks 1893:206. Astrobunus (?) favosum : Weed 1890:914. Trachyrhinus favosus: Weed 1892a:529, 1892b:193, pit. 10, 1893:287; Banks 1894:145, 1901a:675; Scheffer 1906:128; Roewer 1910:266 (in part), 1923:876 (in part), 1957:356; Comstock 1948:72; Katayama and Post 1974:8 (in part). Types.— The female type collected by Prof. F. V. Hayden in Nebraska (Wood, 1871) is presumably lost or destroyed. Attempts to locate Wood’s types in major collections COKENDO LPHER-REVISION OF TRACHYRHINUS 5 (listed in acknowledgments) have failed, and for this reason I am selecting an adult male as the neotype. The neotype was collected at Broken Bow, Custer Co., Nebraska (20 August 1945, M. H. Muma); AMNH. Diagnosis. - Trachyrhinus favosus differs from all other species of Trachyrhinus except T. marmoratus Banks, by having the palpal apophyses 0.07-0.18 in length. T. favosus differs from T. marmoratus by having the ventral surface of the palpal tarsus with many small obtuse tubercles (Figs. 17, 19). Description.— Males (n = 12): Body large, total length 4.84-8.20 (6.04), greatest width 3.00-5.00 (3.94), maximum height 2.37-5.00 (3.46). Ocular tubercle essentially round 0.41-0.56 (0.47), width 0.43-0.53 (0.47); with 4-9 (6) large tubercles on each side. Genital operculum length 1.57-2.80 (2.21), width at base 0.92-1.42 (1.12), width at neck 0.57-0.80 (0.64). Chelicerae black to yellow-brown with brown maculations on first segment dorsally, teeth black. Body ranging from solid black to creamy white with light brown legs that are lightened on the bases of femora. Typically, body light brown to gray-brwon with dark brown maculations; fewer maculations on coxae of legs and ventral surface of body; faint vase-like pattern on dorsum of abdomen. Palpi (Figs. 17-19) black to yellow-brown except for distal portions of femora and patellae and often proximal protions of tibiae dark brown; tarsi with 3-11 (7) ventral denticles, with many small obtuse tubercles. Palpal lengths: femora 0.91-1.30 (1.11), patellae 0.50-0.78 (0.64), patellar apophyses 0.07-0.18 (0.10), tibiae 0.69-0.90 (0.78), tarsi 1.28-1.60 (1.44). Legs black to reddish-brown with bases of femora, dorsa of patellae, disto-dorsal portions of tibiae, and tarsi yellow-brown. Tibiae II with 2-7 (6) pseudosegments. Femora I-IV Fig. 1 2. -Distribution ofT. dicropalpus, T. favosus, T. horneri, and T. mesillensis. 6 THE JOURNAL OF ARACHNOLOGY lengths (respectively): 4.22-7.00 (5.54), 7.75-12.42 (10.04), 4.50-6.84 (5.61), 6.32-9.97 (7.94); tibiae I-IV lengths (respectively): 2.82-5.40 (3.92), 6.83-10.97 (8.83), 3.19-4.85 (4.00), 4.28-6.18 (5.44). Penis as in figures 14-16; length 2.30-3.09 (2.74), width at midshaft 0.10-0.14 (0.12). Females (n = 12): Form and coloration essentially as in males; palpal tarsi without denticles, but with few small obtuse tubercles; total length 4.80-7.91 (6.49), greatest width 3.45-5.01 (4.27), maximum height 2.45-4.62 (3.53). Ocular tubercle, length 0.38-0.52 (0.44), width 0.42-0.52 (0.45); with 3-6 (5) lateral tubercles. Genital operculum length 1.26-1.81 (1.59), width at base 1.20-1.41 (1.43), width at neck 0.54-0.82 (0.70). Palpal lengths: femora 0.87-1.05 (0.95), patellae 0.40-0.58 (0.49), patellar apophyses 0.10-0.18 (0.14), tibiae 0.58-0.74 (0.67), tarsi 1.23-1.45 (1.34). Femora I-IV lengths (respectively): 4.47-5.60 (4.89), 8.21-9.85 (9.10), 4.40-5.45 (4.89), 6.42-7.98 (6.98); tibiae I-IV lengths (respectively): 3.02-3.63 (3.37), 6.63-9.39 (7.98), 3.22-4.50 (3.58), 4.46-5.38 (4.79). Tibiae II with 4-9 (6) pseudosegments. Distribution.— Central United States (Fig. 1 2). Natural History. -Adults of Trachyrhinus favosus are only known to occur from late August to late November. Attempts to maintain specimens in the laboratory have failed. Captive specimens refused cockroaches, moths, and a mixture of cornmeal, yeast, and sugar-water, and generally died within two weeks. A male and a female captured in southern Oklahoma, late November, were guarding 45 white eggs (diameter 0.70-0.82 (0.80), n = 12). The two specimens were under a rock hanging upside down. The eggs were adhering to the rock. Fig. 13.-Distribution of T. marmoratus and T. rectipalpus. COKENDOLPHER-REVISION OF TRACHYRHINUS 7 Specimens examined. -UNITED STATES: Minnesota', Travers Co., Browns Valley, 14 September 1938 (C. E. Mickel), 3 males (ALE): North Dakota', McKenzie Co., T146-R98-S16-P110, 24 August 1976 (collector unknown), 2 males, 3 females (NDSU); Dunn Co., T146-R97-S25-P400, 25 August 1976 (collector unknown), 2 males, 2 females (NDSU): Nebraska', Thayer Co., Gilead, 30 October 1945 (M. H. Muma), 2 males (AMNH); Custer Co., Broken Bow, 20 August 1945 (M. H. Muma), 5 males, 1 female (AMNH): Colorado', Larimer Co., Fort Collins (date and collector unknown), 2 males, 13 females (MCZ); Boulder Co., White Rocks, Boulder Canyon, 30 September 1939 (T. D. A. Cockerell), 1 male (AMNH): Kansas', (specific location, date, and collector unknown), 1 female (AMNH); Sumner Co., South Haven, (date unknown, P. Hayhurst), 1 male (R. V. Chamberlin Coll.- AMNH): Oklahoma : Kay Co., Newkirk, 28 October 1907 (collector unknown), 2 females (Cornell Univ. Coll.-AMNH); Comanche Co., 1 km S Crater Lake, Wichita Mountains Wildlife Refuge, 25 Figs. 1 4-19.-71 favosus : 14-16 penis, 14 dorsal view, 15 lateral view, 16 ventral view; 17-19 male palpus, 17 lateral view, 18 dorsal view, 19 medial view (Scale line for 14-16 =1.0 mm, 17-19 = 3.6 mm). 8 THE JOURNAL OF ARACHNOLOGY November 1977 (J. C. Cokendolpher, D. C. Parmley, F. Bryce), 6 males, 2 females (3 males, 2 females JCC, all others MSU), 23 November 1978 (J. C. Cokendolpher), 2 males (1 male JCC, 1 male MSU); Cotton Co., Red River, S Devol, 15 October 1977 (J. C. Cokendolpher), 1 female (MSU): Texas', Wichita Co., Lake Iowa Park, 3 November 1977 (T. C. Kaspar), 1 male, 1 female (MSU). Trachyrhinus marmoratus Banks Figs. 9-11, 13, 20-25 Trachyrhinus marmoratus Banks 1894:145, 1901a:675; Roewer 1910:268, 1923:876, 1957:356; Goodnight and Goodnight 1942:15, 1946:7; Comstock 1948:72. Trachyrhinus favosus: Banks 1901c:588 (misidentification); Roewer 1910:266 (in part), 1923:876 (in part); Katayama and Post 1974:8 (in part). Trachyrhinus maculatus Roewer 1910:288 ( lapsus calami ). Trachyrhinus sonoranus Chamberlin 1925 : 172. Types.— Only one male Trachyrhinus marmoratus type appears to still exist. I have been unable to locate any of the other males designated by Banks (1894) in major collections (listed in acknowledgments). I designate the male from Sante Fe, Santa Fe Co., New Mexico (pre-1901 collection, T. D. A. Cockerell) in SMF (cat. no. RI/5/29) as the lectotype. This specimen is described and illustrated (Roewer, 1923:876). There are three male and one female topotypes collected by T. D. A. Cockerell (MCZ) that might be part of the original type series, but as the specimens are not labeled types and the series contains a female I am regarding them as a separate collection. Male holotype (CAS, cat. no. 1643) and male paratype (MCZ) of T. sonoranus from Guaymas, Sonora, Mexico, 15 April 1921 (J. C. Chamberlin); both examined. Six paratypes of T. sonoranus from Nogales, Arizona, 4 April 1921 (E. P. Van Duzee); CAS, 1 male and 3 females examined. Diagnosis. - Trachyrhinus marmoratus differs from other species of Trachyrhinus by having an apophysis on the palpal patellae less than 0.18 in length and lacking small obtuse tubercles on ventral surfaces of palpal tarsi. Description.— Males (n = 12): Body large, total length 4.35-7.57 (5.63), greatest width 2.63-4.79 (3.73), maximum height 2.40-4.25 (3.30). Ocular tubercle slightly wider than long, length 0.42-0.56 (0.50), width 0.50-0.58 (0.54); with 7, rarely 8, large lateral tubercles. Genital operculum length 1.50-2.62 (1.98), width at base 0.79-1.42 (1.02), width at neck 0.26-0.70 (0.57). Chelicerae ranging from creamy white to light yellow- brown with brown maculations on first segment dorsally, teeth dark brown. Body colora- tion varies from specimens which are entirely creamy white with few light brown macula- tions on dorsum of abdomen, coxae, and femora to specimens that are reddish-brown with dark brown legs and brown maculations on dorsum, coxae, and venter. Commonly, body yellow-brown with dark brown maculations on dorsum, coxae, and few on venter. Palpi (Figs. 23-25) white to yellow-brown, often with distal portions of femora and patellae brown; tarsi with 2-10 (6) ventral denticles, lacking small obtuse tubercles. Palpal lengths: femora 0.82-1.50 (1.05), patellae 0.42-0.98(0.60), patellar apophyses 0.04-0.10 (0.07), tibiae 0.58-1.04 (0.71), tarsi 1.07-1.55 (1.27). Legs creamy white to dark brown; often with bases of femora, dorsa of patellae, disto-dorsal portions of tibiae, and tarsi yellow-brown. Tibiae II with 3-9 (6) pseudosegments. Femora I-IV lengths (respectively): 3.95-8.52 (5.74), 7.24-11.68 (9.22), 4.20-6.42 (5.37), 5.73-9.38 (7.43); tibiae I-IV lengths (respectively): 2.86-4.39 (3.81), 6.39-10.50 (8.48), 2.82-4.39 (4.00), 3.00-6.22 (4.73). Penis as in figures 20-22; length 2.10-2.85 (2.41), width at midshaft 0.09-0.15 (0.12). COKENDO LPHER-REVISION OF TRACHYRHINUS 9 Females (n = 12): Form and coloration essentially as in males; palpal tarsi bare; total length 4.58-7.84 (5.99), greatest width 3.43-5.62 (4.18), maximum height 2.65-5.00 (3.60). Ocular tubercle length 0.40-0.60 (0.43), width 0.47-0.53 (0.49); with 4-9 (6) lateral tubercles. Genital operculum length 1.30-1.62 (1.49), width at base 1.19-1.54 (1.31), width at neck 0.58-0.81 (0.66). Palpal lengths: femora 0.71-1.02 (0.89), patellae 0.43-0.60 (0.52), patellar apophyses 0.04-0.11 (0.08), tibiae 0.52-0.79 (0.64), tarsi 1.12- 1.42 (1.25). Femora I-IV lengths (respectively): 4.12-6.82 (5.01), 7.18-11 .79 (8.76), 4.12- 6.91 (5.17), 6.01-9.78 (7.19); tibiae I-IV lengths (respectively): 2.63-4.84 (3.54), 6.84-11.00 (8.11), 2.79-4.32 (3.63), 4.22-6.73 (4.99). Tibiae II with 4-11 (6) pseudoseg- ments. Figs. 20-25 .-T. marmoratus : 20-22 penis, 20 dorsal view, 21 lateral view, 22 ventral view; 23-25 male palpus, 23 lateral view, 24 dorsal view, 25 medial view (scale line for 20-22 - 1.0 mm, 23-25 = 3.6 mm). 10 THE JOURNAL OF ARACHNOLOGY Distribution.— Western North America (Fig. 13). Specimens examined (115 males and 107 females, detailed list available from the author). - UNITED STATES: North Dakota', Sargent Co., Slope Co., Dunn Co., McKenzie Co.: Montana', Custer Co.: Nebraska', Thomas Co., Scottsbluff Co., Kimball Co.: Colorado', El Paso Co.: Utah', Washington Co.: California', Riverside Co.: Arizona', Mohave Co., Maricopa Co., Santa Cruz Co., Yuma Co., Pima Co., Cochise Co., Graham Co.: New Mexico', Rio Arriba Co., Santa Fe Co., Bernalillo Co., Hidalgo Co., Eddy Co.: Texas', Brewster Co., Terrell Co., Winkler Co., Potter Co., Lubbock Co.: Oklahoma', Cimarron Co.. MEXICO: Baja California Norte; Sonora; Durango; Chihuahua; Coahuila; Zacatecas; San Luis Potosi. Trachyrhinus dicropalpus, new species Figs. 7,8, 12, 26-31 Mesosoma texanum: Rowland and Reddell 1976: 12 (misidentificatipn). Types.— Male holotype and two female paratypes from Camp Chrysalis, 22 km S Kerrville, Kerr Co., Texas (15 October 1977, D. Holub); holotype FSCA, one paratype MSU, one paratype JCC. Two female paratypes from Ney Cave, Medina Co., Texas (14 April 1972, S. Wiley, T. Mollhagen, and B. Davis); TTU (cat. no. 81). Etymology.— The specific epithet from Greek dicro (meaning forked) and Latin palpus (meaning palp), describing the long apophyses of the palpi. Diagnosis -Trachythinus dicropalpus differs from all other species of Trachyrhinus by having the patellae of palpi with an apophysis which is about one-third length of palpal tibia (0.23 or greater in length). Description.— Male: Body small, total length 4.40, greatest width 2.68, maximum height 2.38. Ocular tubercle essentially round, diameter 0.44, with 7 large tubercles on each side. Genital operculum length 1 .36, width at base 1 .05, width at neck 0.60. Dorsum of body and coxae of legs reddish-brown with dark maculations, venter lighter in color. Chelicerae yellow-brown with black teeth. Palpi (Figs. 29-31) yellow-brown except for distal portions of femora and patellae and proximal portions of tibiae dark brown; tarsi with 3-5 ventral denticles, lacking small obtuse tubercles. Palpal lengths: femur 0.96, patella 0.49, patellar apophysis 0.23, tibia 0.68, tarsus 1.14. Legs yellow-brown except for distal portions of femora, patellae, and tibiae dark brown; disto-dorsal tips of femora, tibiae, and dorsa of tarsi white. Tibiae II with 4 pseudosegments. Femur I-IV lengths (respectively): 5.09, 13.23, 5.07, 6.83; tibia I-IV lengths (respectively): 5.09, 13.23, 5.07, 6.83. Penis as in figures 26-28; length 2.09, width at midshaft 0.10. Females (n = 4): Form and coloration as in male, except dorsum of body brown; total length 4.82-6.40 (5.64), greatest width 3.52-5.10 (4.24), maximum height 3.13-5.00 (3.83). Ocular tubercle, diameter 0.44-0.48 (0.45), with 6, rarely 7, lateral tubercles. Genital operculum length 1 .37-1 .65 (1 .48), width at base 1 .30-1 .44 (1 .38), width at neck 0.60-0.84 (0.71). Palpal lengths: femora 0.94-1.16 (1.03), patellae 0.53-0.72 (0.63), patellar apophyses 0.23-0.29 (0.25), tibiae 0.70-0.79 (0.75), tarsi 1.21-1.28 (1.23). Femora I-IV lengths (respectively): 6.01-7.50 (6.72), 11.01-13.48 (12.14), 5.82-6.92 (6.31), 8.02-10.24 (9.02); tibiae I-IV lengths (respectively): 4.22-5.02 (4.62), 10.05-14.56 (12.07), 4.05-4.60 (4.37), 5.62-6.81 (6.24). Tibiae II with 4-8 (6) pseudoseg- ments. Distribution.— Known only from central Texas (Fig. 12). Natural History. -The specimens from Ney Cave appear to be accidental and are not regarded as part of the cave fauna. The remaining specimens were taken under rocks COKENDOLPHER- REVISION OF TRACHYRHINUS 11 during the daytime. One female captured during April is filled with 145 ova, diameter Q43-0.64 (0.54) for n = 25. Due to the large size and number of ova, it appears oviposition takes place in the spring. Specimens examined. -Only the type series. Figs. 26-31.-7'. dicropalpus : 26-28 penis, 26 dorsal view, 27 lateral view, 28 ventral view; 29-31 male palpus, 29 lateral view, 30 dorsal view, 31 medial view (scale line for 26-28 = 1.0 mm, 29-31 = 3.6 mm). 12 THE JOURNAL OF ARACHNOLOGY Trachyrhinus rectipalpus, new species Figs. 13,32-31 Trachyrhinus sp.: Milstead 1958:445. Trachyrhinus marmoratus: Milstead 1958:445 (misidentification). Types.— Male holotype from Laredo, Webb Co., Texas, 18 December 1939 (F. Norman), AMNH; and 56 paratypes (listed under specimens examined). Etymology.— The specific epithet from Latin recti (meaning straight) and palpus (meaning palp), describing the lack of a palpal apophysis. Diagnosis. - Trachyrhinus rectipalpus differs from all other species of Trachyrhinus , except T. horneri and T. mesillensis, by lacking a palpal apophysis. From T. horneri and T. mesillensis, T. rectipalpus differs by the lengths of femora II: T. horneri mean length 6.04 (n = 3), T. mesillensis 12.19 (n = 2), and T. rectipalpus 9.21 (n = 57). Description.— Males (n = 12): Body relatively small, total length 4.82-6.82 (5.54), greatest width 2.82-4.00 (3.16), maximum height 2.41-3.43 (2.91). Ocular tubercle es- sentially round, diameter 0.40-0.52 (0.45), with 4-8 (5) tubercles on each side. Genital operculum length 1.42-2.21 (1.83), width at base 0.89-1.21 (0.99), width at neck 0.48-0.64 (0.57). Dorsum light brown, venter and coxae lighter; with few dark brown maculations. Chelicerae light yellow-brown with dark brown teeth. Palpi (Figs. 35-37) light brown except bases of femora and tibiae often yellow-brown; tarsi with 4-10 (7) ventral denticles, lacking small obtuse tubercles. Palpal lengths: femora 0.80-1.04 (0.91), patellae 0.52-0.68 (0.61), tibiae 0.55-0.70 (0.64), tarsi 1.18-1.43 (1.33). Legs borwn, often darker on distal portions of femora, patellae, and tibiae. Tibiae II with 3-6 (5) pseudosegments. Femora I-IV lengths (respectively): 4.22-6.87 (5.49), 7.52-12.35 (9.34), 3.81-6.91 (5.52), 5.23-8.22 (7.05); tibiae I-IV lengths (respectively): 3.00-8.79 (4.53), 6.98-11.89 (9.38), 3.18-4.86 (4.11), 4.17-6.78 (5.48). Penis as in figures 32-34: length 2.04-2.45 (2.28), width at midshaft 0.10-0.13 (0.12). Females (n = 12): Form and coloration essentially as in males, total length 5.29-6.72 (5.90), greatest width 3.31-4.21 (3.84), maximum height 2.40-4.02 (3.29). Ocular tubercle, diameter 0.39-0.50 (0.46), with 1-6 (5) lateral tubercles. Genital operculum length 1.20-1.43 (1.33), width at base 1.11-1.48 (1.24), width at neck 0.60-0.72 (0.67). Palpal lengths: femora 0.71-0.91 (0.83), patellae 0.50-0.59 (0.54), tibiae 0.53-0.60 (0.56), tarsi 1.11-1.35 (1.24). Femora I-IV lengths (respectively): 4.37-6.25 (5.03), 8.22- 10.45 (9.05), 4.38-6.01 (4.93), 6.18-7.38 (6.63); tibiae I-IV lengths (respectively): 2.71- 4.26 (3.38), 7.18-9.14 (8.22), 3.26-3.79 (3.46), 4.21-5.23 (4.69). Tibiae II with 4-7 (4) pseudosegments. Distribution.— South Texas in the United States, northern Coahuila and Tamaulipas in Mexico (Fig. 13). Specimens examined.— 56 paratypes- UNITED STATES: Texas', McCulloch Co., 4.8 km E Salt Gap. 15 October 1968 (Perkins), 4 males, 4 females (NDSU); Tom Green Co., 23 October 1968 (Perkins), 9 males, 7 females (NDSU), 2 males, 2 females (JCC), 1 male, 1 female (CAS), 1 male, 1 female (FSCA), 1 male, 1 female (SMF); Bexar Co., San Antonio, 23 August 1940 (L. I. Davis), 1 female (MCZ); Goliad Co., Berclair, 2 January 1952 (Causey), 1 male (MCZ), Goliad, 2 January 1952 (Causey), 2 males, 2 females (MCZ), San Antonio River, 2 January 1952 (Causey), 3 males (MCZ); Nueces Co., Corpus Christi, 1-20 July 1935 (H. C. Sibley, Jr.), 1 female (AMNH); Webb Co., Laredo (date and collector unknown), 1 male (AMNH), 51.2 km E Laredo, 11 November 1934 (Mulaik), 2 males (AMNH); Hidalgo Co., Edinburg, 9 October 1935 (C. Rutherford), 1 male (AMNH), 4 December 1935 (M. Welch), 1 male (AMNH), 5 December 1935 (Mulaik), 1 male (AMNH), 7 December 1935 (C. Rutherford), 1 male (AMNH); Val Verde Co., Lake Walk, 11 August 1966 (T. S. Briggs), 1 female COKENDO LPHER-REVISION OF TRA CHYRHINUS 13 (CAS); Zapata Co., Falcon Reservoir State Park, 12 August 1966 (T. S. Briggs), 1 male, 1 female (CAS). MEXICO: Tamaulipas, 48 km S Reynosa, 24 April 1967 (W. Peck), 1 female (EP): Coahuila, Gloria, 24 August 1947 (W. J. Gertsch), 1 female (AMNH). Figs. 32-37.- T. rectipalpus : 32-34 penis, 32 dorsal view, 33 lateral view, 34 ventral view; 35-37 male palpus, 35 lateral view, 36 dorsal view, 37 medial view (scale line for 32-34 = 1.0 mm, 35-37 = 3.6 mm). 14 THE JOURNAL OF ARACHNOLOGY Trachyrhinus mesillensis, new species Figs. 12, 38-40 Trachyrhinus marmoratus : Banks, 1901b:593 (misidentification). IVpes.-Female holotype and female paratype from Mesilla Valley, Dona Ana Co., New Mexico (N. Banks Coll.-MCZ). Although the specimens are labeled “Mesilla N. Mex.”, Banks (190 lb: 593) reports the specimens are from Las Cruces, September (Cockerell). Mesilla and Las Cruces are only a few km apart and both are in the Mesilla Valley. Etymology.— The specific epithet is an adjectival form derived from the type locality. Diagnosis.— Trachyrhinus mesillensis differs from all other species of Trachyrhinus by lacking patellar apophyses and by having long slender legs (femora II 12.00 or greater in length). Description. -Male: Unknown. Females (measurements of holotype listed first): Body relatively large, robust; total length 6.40-6.08, greatest width 4.07-3.68, maximum height 3.47-3.79. Ocular tubercle slightly longer than wide, length 0.42-0.45, width 0.40-0.43; with 3-5 lateral tubercles. Genital operculum length 1.36-1.41, width at base 1.23-1.30, width at neck 0.70-0.72. Body light yellow-brown with few brown maculations. Chelicerae, palpi, and legs yellow- brown with patellae and distal ends of femora and tibiae of legs darkened. Palpi (Figs. 38-40) lacking apophyses or ventral tubercles on tarsi. Palpal lengths: femora 0.84-?, Figs. 38-40 .-T. mesillensis-. 38-40 female palpus, 38 lateral view, 39 dorsal view, 40 medial view (scale line = 3.6 mm). COKENDO LPHER-REVISION OF TRA CHYRHINUS 15 patellae 0.54-?, tibiae 0.58-?, tarsi 1.30-?. Legs very long and slender, tibiae II with 3, 5 - 5, 7 pseudosegments. Femora I-IV lengths (respectively): 6.39-6.60, 12.00-12.38, 6.31-6.01, 8.75-9.09; tibiae I-IV lengths (respectively): 4.21-4.60, 9.22-11.03, 4.39-4.48, 5.94-6.19. Distribution. -Known only from Mesilla Valley, New Mexico (Fig. 12). Specimens examined. -Only the type specimens. Figs. 41-46. -7! horneri : 41-43 penis, 41 dorsal view, 42 lateral view, 43 ventral view; 44-46 male palpus, 44 lateral view, 45 dorsal view, 46 medial view (scale line for 41-43 = 1.0 mm, 44-46 = 3.6 mm). 16 THE JOURNAL OF ARACHNOLOGY Trachyrhinus horneri, new species Figs. 6, 12,41-46 Types.— Male holotype and two female paratypes from 16 km SW Abilene, Taylor Co., Texas, February 1944 (H. S. Dybas); FMNH, one paratype JCC. Etymology.— The specific epithet is a patronym honoring Dr. Norman V. Florner, who first introduced me to the study of arachnology. Diagnosis.— This species differs from all other species of Trachyrhinus by having short legs (femora II less than 6.5 in length) and lacking palpal apophyses. Females of T. horneri differ from all other species of Trachyrhinus by having 2 pseudoarticulary nodules in femora II. Description.— Male: Body small, length 4.51, greatest width 3.00, maximum height 2.41. Ocular tubercle slightly wider than long, length 0.40, width 0.46; with 3 tubercles on each side. Genital operculum without lateral rows of tri-pointed denticles; length 1 .49, width at base 0.84, width at neck 0.50. Dorsum light brown with few dark brown maculations; ocular tubercle and venter white to light brown with few dark brown macu- lations on venter. Chelicerae light yellow-brown with dark brown teeth. Palpi (Figs. 44-46) light brown except for bases of femora white; tarsi with 5 ventral denticles, lacking small tubercles. Palpal lengths: femur 0.78, patella 0.51, tibia 0.50, tarsus 1.09. Legs light brown; darker on distal portions of femora, dorso-lateral portions of patellae, and laterally on tibiae. Tibiae II with 2 pseudosegments. Femur I-IV lengths (respectively): 3.19, 6.00, 3.41, 4.58; tibia I-IV lengths (respectively): 2.38, 5.62, 2.40, 3.21. Penis as in figures 41-43; length 1.76, width at midshaft 0.12. Females (n = 2): Form essentially as male, coloration lighter, total length 5.10-6.10, greatest width 3.60-3.64, maximum height 3.18-3.68. Ocular tubercle length 0.41-0.50, width 0.43-0.50; with 4-6 lateral tubercles. Genital operculum length 1.35-1.40, width at base 1.28-1.32, width at neck 0.67-0.70; with a few tri-pointed tubercles on lateral margins. Palpal lengths: femora 0.80-1.09, patellae 0.50-0.54, tibiae 0.52-0.54, tarsi 1.13-1.28. Femora I-IV lengths (respectively): 3.38-3.42, 5.94-6.18, 3.41-3.52, 4.28-4.71; tibiae I-IV lengths (respectively): 2.42-2.60, 5.52-5.98, 2.49-2.60, 3.42-3.61. Tibiae II with 4-5 pseudosegments; femora II with 2 pseudoarticulary nodules (Fig. 6). Distribution.— Known only from near Abilene, Texas (Fig. 12). Specimens examined. -Only the type specimens. ACKNOWLEDGMENTS Dr. Clarence J. and Marie L. Goodnight and William A. Shear are thanked for many helpful comments and suggestions. Dr. Oscar F. Francke and Mr. W. David Sissom are gratefully acknowledged for their critical reviews of the manuscript. Dr. Norman V. Horner and the Department of Biology, Midwestern State University provided me with laboratory space and equipment during my studies at that institution. Thanks are also extended to the following individuals and curators for loans and gifts of specimens: Dr. Paul H. Arnaud, Jr. and Dr. David H. Kavanaugh, California Academy of Sciences, San Francisco, California (CAS); Dr. Edward U. Balsbaugh, Jr ., North Dakota State University, Fargo, North Dakota (NDSU); Dr. George W. Byers, Snow Entomological Museum, Lawrence, Kansas (SEM); Dr. R. E. Crabill, Jr., United States National Museum, Washington, D. C. (USNM); Dr. Arlan L. Edgar, Alma College, Alma, Michigan (ALE); Dr. G. B. Edwards, Florida State Collection of Arthropods, Gainesville, Florida (FSCA); Dr. COKENDO LPHER- REVISION OF TRA CHYRHINUS 17 Oscar F. Francke, Texas Tech University, Lubbock, Texas (OFF); Dr. M. Grasshoff, Natur-Museum Senckenberg, Frankfurt, West Germany (SMF); Dr. Norman V. Horner, Midwestern State University, Wichita Falls, Texas (MSU); Dr. Herbert W. Levi, Museum of Comparative Zoology, Cambridge, Massachusetts (MCZ); Dr. William B. Peck, Exline- Peck Collection, Warrensberg, Missouri (EP); Dr. Norman I. Platnick, American Museum of Natural History, New York, New York (AMNH); Dr. William A. Shear, Hampden- Sydney College, Hampden-Sydney, Virginia (WAS), specimens on loan to Shear from University of California at Riverside, California (UCR); Dr. Eric Smith, Field Museum of Natural History, Chicago, Illinois (FMNH); Dr. James R. Zimmerman and Dr. Dave Richman, Museum of Entomology, New Mexico State University. Las Cruces, New Mexico (NMSU). In addition to the above named curators, thanks are also extended to the following curators for searching for type specimens of Trachyrhinus in their respective msueums: Dr. David Barr, Royal Ontario Museum, Toronto, Canada; Dr. Charles D. Dondale, Canadian National Collection of Insects and Arachnids, Ottawa, Canada; Dr. Orsetta Elter, Museo ed Istitutio di Zoologia Sistematica, Torino, Italy; Dr. J. Kekenbosch, Institut Royal des Sciences Naturelles de Belgique, Bruxelles, Belgium; Dr. Daniel Otte, The Academy of Natural Sciences, Philadelphia, Pennsylvania; Dr. Maria Rambla, Instituto de Biologia Aplicada, Barcelona, Spain; Dr. E. Sutter, Naturhistorisches Museum, Basel, Switzerland; Dr. John D. Unzicker, Illinois Natural History Survey, Urbana, Illinois; Dr. George Wallace, Carnegie Museum of Natural History, Pittsburgh, Pennsylvania; Dr. F. R. Wanless and Mr. P. D. Hillyard, British Museum (Natural History), London, England. LITERATURE CITED Banks, N. 1893. The Phalanginae of the United States. Canadian Entomol., 25:205-211. Banks, N. 1894. Notes on Phalangidae. J. New York Entomol. Soc., 2(4): 145-146. Banks, N. 1901a. Synopses of North-America invertebrates. XVI. The Phalangidae. Amer. Nat., 35: 669-679. Banks, N. 1901b. Some Arachnida from New Mexico. Proc. Acad. Nat. Sci. Philadelphia, 53: 568-597. Banks, N. 1901c. Some spiders and other Arachnida from southern Arizona. Proc. United States Natl. Mus., 23: 581-590. Chamberlin, R. V. 1925. Expedition of the California Academy of Sciences to the Gulf of California in 1921, the Phalangidae. Proc. California Acad. Sci., 4th Ser., 14(9): 171-173. Comstock, J. H. 1948. (revised and edited by W. J. Gertsch). The spider book. Cornell University Press. Ithaca, New York, 795 p. Goodnight, C. J. and M. L. Goodnight. 1942. New and little known Phalangida from Mexico. Amer. Mus. Novit., No. 1163, 16 p. Goodnight, C. J. and M. L. Goodnight. 1946. Additional studies on the phalangid fauna of Mexico. 1. Amer. Mus. Novit., No. 1310, 17 p. Katayama, R. W. and R. L. Post. 1974. Phalangida of North Dakota. North Dakota Insects, No. 9, 40 P- Milstead, W. W. 1958. A list of the arthropods found in the stomachs of whiptail lizards from four stations in southwestern Texas. Texas J. Sci., 10(4): 443-446. Roewer, C. F. 1910. Revision der Opiliones Plagiostethi (= Opiliones Palpatores). l.Teil: Familie der Phalangiidae (Subfamilien Gagrellini, Liobunini, Leptobunini). Abhandl. Naturwiss. Ver. Hamburg, 19(4): 1-294. Roewer, C. F. 1923. Die Weberknechte der Erde, Systematische Bearbeitung der bisher bekannten Opiliones. Gustav Fischer, Jena, 1116 p. Roewer, C. F. 1957. Uber Oligolophinae, Caddoinae, Sclerosomatinae, Leiobunianae, Neopilioninae and Leptobuninae (Phalangiidae, Opiliones, Palpatores). Senckenberg. Biol., 38: 323-358. Rowland, J. M. and J. R. Reddell. 1976. Annotated checklist of the arachnid fauna of Texas (exclud- ing Acarida and Araneida). Occ. Pap. Mus. Texas Tech Univ. No. 38, 25 p. 18 THE JOURNAL OF ARACHNOLOGY Scheffer, T. H. 1906. Additions to the list of Kansas Archnida. Trans. Kansas Acad. Sci., 20: 121-130. Underwood, L. M. 1885. A preliminary list of the Arthogastra of North America (excluding Mexico). Canadian Entomol., 17(Phalangiinae): 167-169. Weed, C. M. 1889. A partial bibliography of the Phalangiinae of North America. Bull. Illinois State Lab. Nat. Hist., 3: 99-106. Weed, C. M. 1890. The harvest spiders of North America. Amer. Nat., 24: 914-918. Weed, C. M. 1892a. Notes on harvest-spiders. Amer. Nat., 26: 528-530. Weed, C. M. 1892b. New or little-know North-American harvest spiders. Trans. Amer. Entomol. Soc., 19: 187-194. Weed, C. M. 1893. A synopsis of the harvest spiders (Phalangiidae) of South Dakota. Trans. Amer. Entomol. Soc., 20: 285-292. Wood, H. C. 1871. On the Phalangeae of the United States of America. Comm. Essex Institute, 6: 10-40. Manuscript received October 1979, revised November 1979. Rowland, J. M. and J. R. Reddell 1981. The order Schizomida (Arachnida) in the New World. IV. goodnightorum and briggsi groups and unplaced species (Schizomidae, Schizomus). J. Arachnol., 9:19-46. THE ORDER SCHIZOMIDA (ARACHNIDA) IN THE NEW WORLD. IV. GOODNIGHTORUM AND BRIGGSI GROUPS AND UNPLACED SPECIES (SCHIZOMIDAE: SCHIZOMUS)1 J. Mark Rowland2 and James R. Reddell3 The Museum and Department of Biological Sciences Texas Tech University, Lubbock, Texas 79409 ABSTRACT This is the fourth and final part of a systematic revision of the order Schizomida (Arachnida) in the New World. The goodnightorum and briggsi species groups are revised and four species which cannot be placed in recognized groups are described. The following species are described and assigned to the goodnightorum group: S. goodnightorum (Rowland), S. orthoplax Rowland, S. lanceolatus Rowland, and S. silvino Rowland and Reddell. The following species are described and assigned to the briggsi group: S. pentapeltis (Cook), S. wessoni (Chamberlin), S. borregoensis (Briggs and Horn), S. shoshonensis (Briggs and Horn), S. joshuensis (Rowland), S. briggsi (Rowland), and S. belkini (McDonald and Hogue). The following unplaced species are described: S. troglobius new species, S. infernalis Rowland, Schizomus sp. from Sierra Nevada, Colombia, and S. armasi new species. INTRODUCTION This is the fourth and final part of a revision of the arachnids of the order Schizomida in the New World. Previous reports have included the family Protoschizomidae and the Schizomus dumitrescoae group of the family Schizomidae (Rowland and Reddell 1979a), the S. simonis and S. brasiliensis groups (Rowland and Reddell 1979b), and the S. mexicanus and S. pecki groups (Rowland and Reddell 1980). The present report includes the S. goodnightorum group from southern Mexico and Guatemala and the S. briggsi group from the southwestern United States; also included are descriptions of four species which cannot be placed in any of the recognized species groups. Table 1 may be used to compare the species groups included here with the remaining groups of New World schizomids. A discussion of the zoogeography and phylogeny of the order in the New World is reserved for a future publication. 1 Supported in part by The Museum, Texas Tech University, and by a Society of Sigma Xi Grant-in-Aid of Research to the senior author. 2 Present address: Department of Pharmacology and Therapeutics, Texas Tech University School of Medicine, Lubbock, Texas 79409. 3 Present address: The Texas Memorial Museum, The University of Texas at Austin, 24th & Trinity, Austin, Texas 78705. 20 THE JOURNAL OF ARACHNOLOGY The present study is based to a large extent on a dissertation prepared by the senior author at Texas Tech University, Lubbock, Texas (Rowland 1975a). Family Schizomidae GOODNIGHTOR UM GROUP Description.— Members of this group are characterized by small to large size (0.89-1.42 mm carapacial length). The color is brownish. The eyespots are distinct to indistinct, and are with irregular to round margins. The carapace has three or four pairs of dorsal and Tab! e 1 . --Compari sons of the New World species groups of the genus Schizomus. See Rowland and Reddell (1979) for explanation of characters. CHARACTER dumitres- coae simon is brasi 1 - iensis mexi- canus pecki goodni- ghtorum briggsi DORSAL SETAE 2-3 2-3 3-4 2-3 2-3 3-4 3-4 METAPEL- TIDIUM entire enti re spl it or entire entire entire entire spl it or enti re COLOR brown or green brown or green brown or green brown or green brown brown brown or green SPERMA- THECAE M < L M < L M = L M > L M > L M > L mul tiple ART. FEM. FLAGELLUM 4 4 3 3 3 3 4 CARAPACE LENGTH .96-1.37 1 .07-1.34 .91-1.48 .98-1.37 1.31-1.74 .89-1.42 1.18-1.52 ABDOMINAL ELONGATION none present none none none present none or present ABDOMINAL PROCESS present present present absent absent absent present PEDIPALPAL DIMORPHISM si ight to strong none slight to strong none to strong none none none to strong SHAPE MALE FLAGELLUM bulbous long bul bous bulbous bul bous long long or bul bous ROWLAND AND REDDELL-NEW WORLD SCHIZOMIDA IV 21 two apical setae. Abdominal segments VII or VIII to XII may be slightly to extremely attenuated in the males. No posterodorsal abdominal process is evident. The male flagel- lum is always elongate, narrow, and apically slender, and has either double or a single median pit. The female flagellum is moderate in length (0.22-0.31 mm) and is composed of three articles. The female spermathecae are characterized by the median pair being much longer than the laterals, with small or no terminal bulbs, and with slight scleroti- zation. The pedipalps are not sexually dimorphic. Distribution.— Mexico: Veracruz, Chiapas, Yucatan. Guatemala. Remarks.— See Table 2 for comparisons of the species in th e goodnightorum group. Subordinate taxa.-S. goodnightorum, S. orthoplax, S. lanceolatus, S. silvino. Schizomus goodnightorum (Rowland) Figs. 1,5-6, 8-9, 14 Heteroschizomus goodnightorum Rowland 1973a: 2-6; Rowland 1973b: 202. Schizomus goodnightorum: Rowland and Reddell 1977: 83, 99, 100. Description.— Male. Color brownish. Carapace with three pairs of dorsal and two apical setae. Eyespots indistinct. Anterior sternum with 10 bifid setae. Abdominal terga I-VII Fig. 1.— Map showing distribution of schizomids of th q goodnightorum group: 1, S. lanceolatus’, 2, S. orthoplax’, 3, S. silvino’, 4, S. goodnightorum. 22 THE JOURNAL OF ARACHNOLOGY with two setae, terga VIII-IX with four setae, segments VII-XII extremely elongated, segment XII with no evidence of posterodorsal process. Vestigial stigmata lighter than sterna. Flagellum long and extremely narrow, with a median pit. Pedipalpal trochanter produced distally; tarsal-basitarsal spurs about 1/5, claw about 1/3 length of tarsus- basitarsus. Leg segment measurements given in Table 3. Female. Flagellum composed of three articles. Spermathecae with medians several times longer than laterals, both divergent, the laterals thicker than medians, the latter terminating in slightly sclerotized, vaguely defined bulbs. Type data.-Holotype male and paratype male taken at Chichen Itza, Yucatan, Mexico, June 1948 (C. Goodnight) (AMNH, examined). Comparisons.— S. goodnightorum has a much greater elongation of distal abdominal segments in the male, and one fewer pair of dorsal carapacial setae than other members of this group. Males of the other species have a pair of dorsal flagellar depressions, while S. goodnightorum has a single depression. The male of this species is further distinguishable from other males with any degree of attenuation of distal abdominal segments outside the goodnightorum group by its lack of a posterodorsal abdominal process. The median spermathecae of S. goodnightorum are much longer than those of S. silvino, the only other species in this group known by females. Distribution.— Known only from the type locality. Remarks.— The placement of this species in the genus Schizomus was, in part, neces- sitated by the discovery of several other species with a distinct, although less, attenuation of the distal abdominal segments. Since the original description a female, undoubtedly referable to this species, has also become available for study. It has thus become apparent that the extreme attenuation of the abdomen is a sexually dimorphic feature and that the females are without pecularities which would separate them from other species in the genus Schizomus. Variation. -The male paratype is smaller than the holotype and exhibits a lesser degree of abdominal attenuation. As was noted in the original description the paratype contained a nematode parasite in the abdominal cavity which may account for its smaller size, but variation of this magnitude is not unusual in other species. Additional record.- Yucatan: Chichen Itza, June 1948 (C. and M. Goodnight), one female (AMNH, examined). Schizomus orthoplax Rowland Figs. 1,3,7, 11 Schizomus orthoplax Rowland 1973a: 6, 10-13; Rowland 1973c: 135; Rowland and Reddell 1977: 99, 100. Description. -Male. Color brownish. Carapace with four pairs of dorsal and two apical setae. Eyespots distinctly round. Anterior sternum with 10 bifid setae. Abdominal terga I-VII with two setae, terga VIII-IX with four setae, segment XII with no evidence of posterodorsal process. Vestigial stigmata darker than sterna. Flagellum lanceolate, with a pair of median pits on otherwise flat dorsal surface. Pedipalpal trochanter slightly pro- duced distally; tarsal-basitarsal spurs about 1/4, claw about 1/2 length of tarsus-basitarsus. Tarsal-basitarsal segments of leg I of the following approximate proportions: 35-4-7-6-7-7-16. Other leg segment measurements given in Table 3. ROWLAND AND REDDELL-NEW WORLD SCHIZOMIDA IV 23 Figs. 2-7. -Parts of male schizomids of the go odnightorum group: 2-5, dorsal views of flagella: 2, S. lanceolatus ; 3, S. orthoplax ; 4, S. silvino, 5, S. goodnightorunv, 6, 7, lateral views of right pedipalps: 6, S', goodnightorunv, 1 , S. orthoplax. 24 THE JOURNAL OF ARACHNOLOGY Table 2. -Comparisons of the members of the goodnightorum group. See the introduction to Rowland and Reddell (1979a) for discussion of characters. CHARACTER goocini- ghtorum orth- oplax lance- olatus silvino DORSAL SETAE 3 4 4 4 STERNAL SETAE 10 10 11 10 EYESPOTS indis- tinct distinct distinct distinct SPERMA- THECAE M 5X L ? ? M 2X L CARAPACE LENGTH .89 1.04 1.42 1.05 LENGTH FEM. FLAGELLUM .22 ? ? .31 ABDOMINAL ELONGATION 7-12 8-12 7-12 8-12 PIT MALE FLAGELLUM single double double double Female unknown. Type data.— Male holotype taken at Finca Cuauhtemoc, Chiapas, Mexico, 8 May 1950 (C. and M. Goodnight) (AMNH, examined). Comparisons.— This species is most similar to S. silvino and S. lanceolatus. They are best distinguished on the basis of the morphology of the male flagellum. The carapacial lengths also differ among these species. The flagellum is by far the longest in S. lanceolatus and by far the shortest in S. orthoplax and is apparently intermediate in S. silvino . See also under S. goodnightorum. Distribution.— Known only from the type locality. Additional record. -Chiapas: Finca Cuauhtemoc, 8 June 1950 (C. and M. Goodnight), two immatures (AMNH). Schizomus lanceolatus Rowland Figs. 1-2, 10 Schizomus lanceolatus Rowland 1975b: 7, 15-16, 17; Rowland and Reddell 1977: 80, 86, 99, 100. ROWLAND AND REDDELL-NEW WORLD SCHIZOMIDA IV 25 Figs. 8-14. -Parts of schizomids of the goodnightorum group: 8, dorsal view of male S. good- nightorum, legs and pedipalps past the trochanter omitted; 9-12, lateral views of male flagella: 9, S. goodnightorum ; 10, S. lanceolatus; 11, S. orthoplax’, 12, S. silvino ; 13, 14, female spermathecae: 13, S. silvino-, 14, S. goodnightorum. 26 THE JOURNAL OF ARACHNOLOGY Description.— Male. Color brownish. Carapace with four pairs of dorsal and two apical setae. Eyespots distinct, irregular. Anterior sternum with 1 1 bifid setae. Abdominal terga I-VII with two setae, terga VIII-IX with four setae, abdominal segments VII-XII attenuate, segment XII with no evidence of posterodorsal process. Vestigial stigmata slightly darker than sterna. Flagellum lanceolate, with pair of median pits on otherwise flat dorsal surface. Pedipalpal trochanter produced distally; tarsal-basitarsal spurs about 1/4, claw about 1/2 length of tarsus-basitarsus. Tarsal-basitarsal segments of leg I of the following approximate proportions: 57-9-9-9-10-9-21. Other leg segment measurements given in Table 3. ' Female unknown. Type data.-Holotype male taken in Cueva del Diablo, near Ciudad Mendoza, Veracruz, Mexico, 7 March 1973 (J. Reddell) (AMNH, examined). Comparisons.-See under S. goodnightorum and S. orthoplax. Distribution.— Known only from the type locality. Remarks.— This dark species has distinct eyespots and is probably a facultative troglophile. It was collected beneath a rock in the dark zone of Cueva del Diablo. Schizomus silvino Rowland and Reddell Figs. 1,4, 12-13 Schizomus silvino Rowland and Reddell 1977: 80, 81, 86, 96-97 , 100. Description.— Male. Color brownish, pale. Carapace with four pairs of dorsal and two apical setae. Eyespots distinct, irregular. Anterior sternum with 10 bifid setae. Abdominal terga I-VII with two setae, terga VIII-IX with four setae, segments VIII-XII attenuate, segment XII with no evidence of posterodorsal process. Vestigial stigmata lighter than sterna. Flagellum extremely elongate, with a pair of median pits on otherwise flat dorsal surface. Pedipalpal trochanter produced distally; tarsal-basitarsal spurs about 1/5, claw about 2/5 length of tarsus-basitarsus. Tarsal-basitarsal segments of leg I of the following approximate proportions: 44-6-8-7-8-8-18. Other leg segment measurements given in Table 3. Female. Flagellum composed of three articles. Median spermathecae two or three times longer than laterals, with both very slightly divergent, but neither expanded distally; medians sclerotized slightly on apical half. Type data.— Holotype male, allotype female, and three male, six female, and six immature paratypes taken in Gruta de Silvino, 34 km W Puerto Barrios, Izabal, Guatemala, between 20-22 August 1969 (S. and J. Peck) (AMNH, examined). Comparisons.— See under S. goodnightorum and S. orthoplax. Distribution.— Known only from the type locality. Remarks.— Although this species is paler than most, the presence of distinct eyespots indicates that it is a facultative troglophile. BRIGGSI GROUP Description.— The members of this group are characterized by moderate to large size (1.17-1.50 mm carapacial length). The color is usually brownish but may be greenish. The eyespots are indistinct or distinct. The carapace has one or three to four pairs of dorsal and three apical setae. The abdomen is attenuated in males of one species, but not ROWLAND AND REDDELL-NEW WORLD SCHIZOMIDA IV 27 Table 3. -Measurements (mm) of the species of the goodnightorum group: 1, two males, S. good- nightorunv, 2, one female, S. goodnightorum ; 3, one male, S. orthoplax; 4, one male, S. lanceolatus", 5 , two males, S. silvino\ 6, two females, S. silvino. Except as otherwise noted all measurements are of lengths. 1 2 3 4 5 6 Carapace 1.04-1.12 0.89 1.04 1.42 1.11-1.14 1.00-1.05 Flagellum Length 1.04 0.22 0.57 0.99 0.70-0.79 0.29-0.31 Width 0.22 - 0.19 0.27 0.20-0.20 _ Leg I Femur 1.33-1.67 0.84 1.02 1.80 1.45-1.62 1.05-1.14 Patella 1.55 0.99 1.24 2.34 1.80-2.01 1.26-1.36 Tibia 1.15 0.72 0.85 1.79 1.03-1.36 0.86-0.91 Tarsus-Basitarsus - 0.60 0.82 1.26 0.97-0.99 0.76-0.77 Leg II Femur 0.83-0.93 0.58 0.60 1.15 0.82-0.92 0.71-0.75 Patella 0.40-0.52 0.33 0.31 0.55 0.48-0.51 0.40-0.41 Tibia 0.56-0.63 0.32 0.38 0.82 0.53-0.59 0.40-0.45 Basitarsus 0.51-0.45 0.30 0.38 0.67 0.49-0.53 0.38-0.43 Leg III Femur 0.77 0.51 0.53 1.01 0.71-0.79 0.61-0.65 Patella 0.30 0.22 0.20 0.50 0.33-0.36 0.29-0.31 Tibia 0.41 0.26 0.30 0.57 0.41-0.44 0.34-0.36 Basitarsus 0.50 0.33 0.37 0.71 0.50-0.55 0.40-0.47 Leg IV Femur 1.16-1.35 0.86 0.94 1.51 1.17-1.31 1.01-1.07 Patella 0.45-0.50 0.40 0.37 0.65 0.51-0.52 0.44-0.46 Tibia 0.80-0.95 0.60 0.64 1.07 0.77-0.86 0.66-0.71 Basitarsus 0.70-0.81 0.50 0.51 0.92 0.70-0.77 0.58-0.64 in others. The males have a posterodorsal abdominal process which is always pointed apically. The flagellum is usually globose, but is long and thin in one species, and usually bears some dorsal modifications. The female flagellum is composed of four articles and is long (0.34-0.54 mm). The female spermathecae are characterized by three, four, or several pairs of short, broad lobes which may or may not be sclerotized apically. The pedipalps may be slightly to extremely or not sexually dimorphic. The trochanter is often produced distally, and the femur, patella, and tibia are often elongated. The tibia in males of all species has either a series of heavy spines or a well-developed spur apposible to the tarsus-basitarsus. Distribution.— United States: California, Arizona. Remarks.— See Table 4 for comparisons of the species in the briggsi group. Subordinate taxa —Pentapeltis complex: S. pentapeltis; briggsi complex: S. wessoni, S. borregoensis, S. shoshonensis, S. joshuensis, S. briggsi, S. belkini. Schizomus pentapeltis (Cook) Figs. 15,23,27,38 Hubbardia pentapeltis Cook 1899: 25 3-354, 261. Trithyreus pentapeltis : Banks 1900: 422; Hansen and Sorensen 1905: 4, 44, 70; Moles 1917: 1-7; Moles 1921: 11; Kishida 1930: 18; Hilton 1932: 33-34, 45-46; Giltay 1935: 8; Werner 1935: 469; Gertsch 1940: 1; Takashima 1943: 96; Comstock 1948: 18; Kraus 1957: 245; McDonald and 28 THE JOURNAL OF ARACHNOLOGY Hogue 1957: 1, 6-7; Briggs and Horn 1966: 270, 273, 274; Horn 1967: 218, 220; Rowland 1971: 304; Rowland 1972a: 69-74; Rowland 1972b: 5, 8; Rowland 1972c: 153, 155, 156, 159; Rowland 1973b: 195, 196, 202. Schizomus pentapeltis: Mello Leitao 1931: 18; Rowland and Reddell 1979a: 162- Description (of a male and female topotype).— Male. Color brownish. Carapace with three pairs of dorsal and three apical setae. Metapeltidium split. Eyespots distinct, irregular. Anterior sternum with 13 bifid setae. Abdominal terga I-IV with two setae, V-VI with four setae, IX with two setae, VII- VIII with six setae, segment XII with acutely produced, well-developed posterodorsal process. Abdominal segments VII-XII attenuated. Vestigial stigmata darker than sterna. Flagellum long, lanceolate, with a pair of median depressions. Pedipalpal trochanter produced distally; tarsal-basitarsal spurs about 1/5, claw about 1/2 length of tarsus-basitarsus. Tarsal-basitarsal segments of leg I of the following approximate proportions: 63-8-12-12-12-11-28. Other leg segment measure- ments given in Table 5. Female. Flagellum composed of four articles. Spermathecae with several small, closely associated lobes, apically sclerotized in localized areas. Type data.-Two male and one female cotypes taken at Palm Canyon, near Palm Springs, Riverside County, California, 13 February and 6 March 1897 (Hubbard) (USNM, examined). Comparisons.— S', pentapeltis is easily distinguished from the other species of the goup by the elongate flagellum and abdomen of the male. The spermathecal morphology is similar to that of other species, but is characterized by lightly sclerotized apical tubercles. The male pedipalps are not sexually dimorphic as in other Californian species. Fig. 15. -Map showing distribution of schizomids of the briggsi group: 1, S. briggsi; 2, S. shoshonensis ; 3, S. belkinv, 4, S. pentapeltis’, 5, S. joshuensis; 6, S. borregoensis; 7 ,S. wessoni. ROWLAND AND REDDELL-NEW WORLD SCHIZOMIDA IV 29 Table 4. -Comparisons of the members of the briggsi group. See the introduction to Rowland and Reddell (1979a) for discussion of characters. CHARACTER penta- pel tis wessoni borrego- ensis shoshon- ensis joshu- ensis briggsi bel kini DORSAL SETAE 3-4 3 3 1 3 3 3 STERNAL SETAE 13 14 13 11 13 13 13 COLOR brown brown brown brown brown green green MALE PEDIPALP short short short long long long long SPERMA- THECAE several 3 several ? 4 4 4 CARAPACE LENGTH 1.50 1.30 1.43 1.21 1.41 1.29 1.17 LENGTH FEM. FLAGELLUM .54 ? .43 .34 .40 .39 .37 ABDOMINAL ELONGATION 7-12 none none none none none none SHAPE MALE FLAGELLUM elongate tril- obate penta- gonal trian- gular hexa- gonal hexa- gonal hexa- gonal PIT MALE FLAGELLUM double absent absent double double singl e double Distribution.— This species is known from the coastal foothills of southernmost California west to the coast and north into the San Jacinto Mountains, and Los Angeles Basin. An apparently isolated population occurs at an oasis in Palm Canyon on the northeast slope of the San Jacinto Mountains. Remarks.— This, the first schizomid described from the New World, has been found to be widespread in California and may occur in Baja California, Mexico, as well. Rowland (1972a) reported the brooding habits and early development of this species. Variation.— As in other species with an elongated abdomen in the males, the abdomen and flagellum seem to vary considerably in the extent to which they are attenuated. Rarely specimens occur which have setational differences from the above description. Four pairs of dorsal carapacial setae may occur, as well as may additional abdominal tergal setae. 30 THE JOURNAL OF ARACHNOLOGY Additional records. -California: Riverside County: Andreas Canyon, 3 March 1956 (V. Roth), one male, one female, two immatures (AMNH), 25-27 March 1960 (W. Gertsch, W. Ivie, Schrammel, V. Roth), six males, 25 females, 16 immatures (AMNH), 6 April 1966 (T. Briggs, K. Horn), four immatures (CAS), 10 January 1971 (J. Rowland), one male, five females, five immatures (TTU), 14 January 1971 (J. Rowland), two males, six females, three immatures (MCZ), 23 March 1971 (J. Rowland), four females, four immatures (LACM); Snow Creek Canyon, 20 March 1954 (collector unknown), three females, one immature (AMNH), 13 April 1955 (J. Belkin), three females, eight immatures (AMNH); Riverside, 26 March 1960 (V. Roth), one male, one female (AMNH); Citrus Experiment Station, Riverside, 21 February 1957 (E. Schlinger), one male (AMNH); near Citrus Experiment Station, Riverside, 1 December 1925 (J. Chamberlin), two males, one female (AMNH); Winchester, 22 January 1967 (W. Icenogle), one male (LACM), 24 January 1971 (W. Icenogle), one male, three females, one immature (LACM), 7 February 1971 (W. Icenogle), one female (TTU); Orange County: 11.4 mi. SW Lower San Juan Camp, Cleveland National Forest, oak litter, 20 December 1966 (A. Jung, D. Owyang, K. Horn), one male, five females, five immatures (CAS); San Clemente, 27 December 1966 (W. Lum), one immature (CAS); San Diego County: San Diego, March 1970 (B. Kaston), two males, one immature (LACM); El Cajon, 2 March 1969 (collector unknown), one male, two females (LACM), 1970 (S. Lewis), two males, one female (LACM); Dripping Springs, near Vail Lake, 6 March 1971 (J. Rowland), two males, four females, one immature (AMNH). Schizomus wessoni (Chamberlin) Figs, 15,22, 26,31 Trithyreus wessoni Chamberlin 1939: 123-124; Gertsch 1940: 1; Takashima 1943: 97; Kraus 1957: 245; McDonald and Hogue 1957: 1, 6; Rowland 1971: 304; Rowland 1972c: 154. Schizomus wessoni : Rowland and Reddell 1979a: 162. Description. -Male. Color brownish. Carapace with three pairs of dorsal and three apical setae. Eyespots small, indistinct. Anterior sternum with 14 bifid setae. Abdominal terga I-VII with two setae, terga VIII-IX with four setae, segment XII with acutely produced, well-developed posterodorsal process. Vestigial stigmata darker than sterna. Flagellum trilobate, the median lobe arising above the basal two. Pedipalpal trochanter produced distally; thick spines from tibia appose tarsus-basitarsus; tarsal-basitarsal spurs about 1/6, claw about 1/2 length of tarsus-basitarsus. Tarsal-basitarsal segments of leg I of the follow approximate proportions: 37-6-7-7-8-9-21. Other leg segment measurements given in Table 5. Female. Flagellum composed of four articles. Spermathecae composed of three pairs of small, broad lobes, each with minute terminal elevations, but with no special scleroti- zation. Type data.— Male holotype taken “under a stone shaded by small bushes and trees growing along the Santa Cruz River” near Tucson, Arizona, April 1938 (R. Wesson) (AMNH, examined). Comparisons.— Males may be distinguished from all other members of the briggsi group in having a distinctly trilobate flagellum. The pedipalps are very similar in development to those of S. borregoensis , especially in the development of the mesal tibial spurs which appose the tarsus-basitarsus. The male flagellum of S. borregoensis is distinctly pentagonal rather than trilobed as in S. wessoni. The presence of only three pairs of spermathecal lobes in S. wessoni serves to distinguish females from other species of the group. Distribution.— Known only from near Tucson, Arizona. Remarks.— Long-term drying of the Santa Cruz River due to agricultural activities upstream from Tucson may have totally eliminated S. wessoni from the type locality and ROWLAND AND REDDELL-NEW WORLD SCHIZOMIDA IV 31 Figs. 16-21. -Male flagella of the briggsi group: 16-20, dorsal views: 16, S. belkini; 17, S. foshuensis; 18, S. briggsi; 19, S. shoshonensis; 20, S. borregoensis; 21, lateral view of S. briggsi. 32 THE JOURNAL OF ARACHNOLOGY Table 5. -Measurements (mm) of four species of the briggsi group: 1, three males, S. pentapeltis ; 2, three females, S. pentapeltis ; 3, one male, S. wessoni; 4, one female, S. wessoni ; 5, two males, S. borregoensis ; 6, one female, S. borregoensis’, 7, one male S. shoshonensis’, 8, one female, S. shosho- nensis. Except as otherwise noted all measurements are of lengths. 1 2 3 4 5 6 7 8 Carapace 1.49-1.54 1.42-1.59 L30 L24 1.38-1.46 1.40 1.25 1.18 Flagellum Length 1.39-1.54 0.52-0.57 0.75 - 0.59-0.62 0.43 0.62 0.34 Width 0.45-0.46 - 0.66 - 0.61-0.62 - 0.69 — Leg I Femur 2.05-2.31 1.41-1.63 1.36 - 1.82-1.83 1.39 1.57 1.23 Patella 2.56-3.28 1.68-1.95 1.60 - 2.25-2.32 1.65 1.85 1.42 Tibia 2.04-2.39 1.36-1.57 - - 1.74-1.77 1.22 1.48 1.11 Tarsus-Basitarsus 1.30-1.59 1.08-1.18 - - 1.19-1.30 1.14 1.14 0.99 Leg II Femur 1.18-1.37 1.00-1.15 0.96 0.96 1.14-1.17 1.00 1.02 0.88 Patella 0.69-0.78 0.55-0.67 0.51 0.53 0.66-0.68 0.56 0.59 0.51 Tibia 0.81-0.93 0.68-0.72 0.64 0.64 0.84-0.87 0.66 0.69 0.56 Basitarsus 0.69-0.78 0.55-0.65 0.54 0.56 0.65-0.67 0.55 0.62 0.50 Leg III Femur 1.05-1.15 0.90-1.09 0.85 0.90 1.03-1.07 0.91 0.92 0.79 Patella 0.51-0.59 0.46-0.51 0.39 0.45 0.51-0.52 0.46 0.45 0.39 Tibia 0.65-0.75 0.57-0.64 0.55 0.55 0.70-0.72 0.56 0.57 0.47 Basitarsus 0.75-0.85 0.64-0.70 0.59 0.60 0.71-0.74 0.61 0.68 0.54 Leg IV Femur 1.53-1.80 1.33-1.51 1.21 1.31 1.54-1.59 1.34 1.39 1.21 Patella 0.74-0.79 0.65-0.73 0.55 0.60 0.69-0.70 0.65 0.61 0.57 Tibia 1.15-1.30 0.99-1.14 0.94 0.93 1.18-1.22 1.02 1.01 0.85 Basitarsus 1.01-1.15 0.86-1.00 0.79 0.80 0.96-0.98 0.83 0.91 0.76 perhaps altogether. A relict population 20 miles south of Ajo, Arizona, may represent an earlier isolated relative of this species. Additional record. -Arizona: Tucson, date unknown (Griswold), one female (AMNH, examined). Schizomus borregoensis (Briggs and Horn) Figs. 15,20,25,28,34-35 Trithyreus borregoensis Briggs and Horn 1966: 270-273; Rowland 1971: 304, 308-309; Rowland 1972b: 5, 8; Rowland 1972c: 153, 155, 156, 159. Schizomus borregoensis : Rowland and Reddell 1979a: 163. Description. -Male. Color brownish. Carapace with three pairs of dorsal and three apical setae. Eyespots indistinct. Anterior sternum with 13 bifid setae. Abdominal terga I-V with two setae, terga VI-IX with four setae, segment XII with acutely produced, well-developed posterodorsal process. Vestigial stigmata darker than sterna. Flagellum pentagonal, with no dorsal relief. Pedipalpal trochanter produced distally; thick spines from tibia appose tarsus-basitarsus; tarsal-basitarsal spurs about 1/6, claw about 1/3 length of tarsus-basitarsus. Tarsal-basitarsal segments of leg I of the following approxi- mate proportions: 51-10-10-10-12-1 1-24. Other leg segment measurements given in Table 5. ROWLAND AND REDDELL-NEW WORLD SCHIZOMIDA IV 33 Figs. 22-27. -Parts of schizomids of the briggsi group: 22-26, male flagella: 22, 23, dorsal views: 22, S. wessoni; 23, S. pentapeltis’, 24-26, lateral views: 24, S', shoshonensis ; 25, S. borregoensis', 26, S. wessoni; 27, female spermathecae of S. pentapeltis. 34 THE JOURNAL OF ARACHNOLOGY Female. Flagellum composed of four articles. Pedipalpal trochanter not as acutely produced. Spermathecae composed of several pairs of lobes which are apically elaborated with many minute terminal elevations, no special sclerotization. Type data.-Male holotype, female allotype, and immature paratype taken “under rocks in palm and sycamore debris near stream” in Borrego Palm Canyon, Anza Borrego State Park, San Diego County, California, 4 April 1966 (T. Briggs, K. Horn) (CAS, examined). Comparisons.— This species does not have the extremely dimorphic pedipalps seen in other California species with the exception of S. pentapeltis. The flagellum completely lacks any dorsal relief in contrast to other briggsi group members. See also under S. wessoni. Distribution.— Known only from the type locality. Additional record. -California: San Diego County: Borrego Palm Canyon, 12 January 1971 (J. Rowland, T. Moisi), two males, one female (AMNH). Schizomus shoshonensis (Briggs and Horn) Figs. 15, 19,24 Trithyreus shoshonensis Briggs and Horn 1972: 1-7. Schizomus shoshonensis: Rowland and Reddell 1977: 80; Rowland and Reddell, 1979a: 163. Figs. 28-33. -Female spermathecae of the briggsi group: 28, S. borregoensis ; 29, S. joshuensis’, 30, S. briggsi ; 31, S', wessoni; 32, 33, S. belkini: 32, from the type locality; 33, from Santa Cruz Island. ROWLAND AND REDDELL-NEW WORLD SCHIZOMIDA IV 35 Description.-Male. Color brownish. Carapace with one pair of dorsal and three apical setae. Abdominal terga I- VII with two setae, terga VIII-IX with four setae, segment XII with well-developed, acute posterodorsal process. Vestigial stigmata slightly darker than sterna. Flagellum heart shaped, with pair of paramedial elevations separated by a wide median depression which deepens and widens distally. Pedipalpal trochanter produced slightly distally; all segments elongate; tibia with a spur apposible to tarsus-basitarsus; tarsal-basitarsal spurs about 1/6, claw about 2/5 length of tarsus-basitarsus. Tarsal- basitarsal segments of leg I of the following approximate proportions: 48-7-10-9-9-9-23. Other leg segment measurements given in Table 5. Female. Carapace with three pairs of dorsal setae. Flagellum composed of four articles. Spermathecae not dissected, but appearing typical for the group as seen through the lightly sclerotized genital sternum. Type data.— Male holotype and female allotype taken in Upper Shsoshone Cave, near Shoshone, Inyo County, California, 28 December 1971 (W. Rauscher, E. Fogarino, T. Briggs) (CAS, examined). Comparisons.-This is the only species in the briggsi group with only one pair of dorsal carapacial setae in the male. Its closest relatives, with which it bears obvious similarities, are more western. Also see under S. belkini. Distribution. -Known only from the type locality. Remarks.— This species is unquestionably troglobitic, as was convincingly pointed out by Briggs and Horn (1972). It is likely that other populations of this or related species may occur in nearby caves. The presence of eyespots indicates that it is probably a recent relict. Schizomus joshuensis (Rowland) Figs. 15, 17,29,40 Trithyreus joshuensis Rowland 1971: 304-308; Briggs and Horn 1972: 2; Rowland 1972b: 3, 4,5, 7; Rowland 1972c: 153, 155, 156, 159. Schizomus joshuensis: Rowland and Reddell 1979a: 163. Description.— Male. Color brownish. Carapace with three pairs of dorsal and three apical setae. Eyespots indistinct. Anterior sternum with 13 bifid setae. Abdominal terga I- VII with two setae, terga VIII-IX with four setae, segment XII with acutely produced well-developed posterodorsal process. Vestigial stigmata darker than sterna. Flagellum roughly hexagonal, with pair of distinct median depressions flanking slight median ridge. Pedipalpal trochanter, femur, patella, and tibia extremely elongate; the tibia with a mesal spur apposible to the tarsus-basitarsus; tarsal-basitarsal spurs about 1/6, claw about 1/4 length of tarsus-basitarsus. Tarsal-basitarsal segments of leg I of the following approximate proportions: 51-7-9-9-10-11-25. Other leg segment measurements given in Table 6. Female. Flagellum composed of four articles. Spermathecae composed of four pairs of small, broad lobes, each with minute terminal elaborations, no special sclerotization. Type data.— Holotype male and allotype female taken at Forty-nine Palms, Joshua Tree National Monument, San Bernardino County, California, 20 February 1970 (J. Rowland and D. Harris) (AMNH, examined); paratype male and female taken at Forty- nine Palms, 22 February 1970 (J. and C. Rowland) (AMNH, examined); paratype male and female taken at Forty-nine Palms, 30 December 1970 (J. Rowland and P. Brashier) (AMNH, examined). 36 THE JOURNAL OF ARACHNOLOGY Table 6. -Measurements (mm) of three species of the briggsi group: 1, four males, S. foshuensis; 2, four females, S. foshuensis', 3, five males, S. briggsi', 4, five females, S. briggsi; 5, four males, S. belkini; 6, four females, S. belkini. Except as otherwise noted all measurements are of lengths. 1 2 3 4 5 6 Carapace 1.36-1.48 1.39-1.47 1.22-1.32 1.22-1.40 1.13-1.25 1.11-1.24 Flagellum Length 0.63-0.65 0.38-0.44 0.55-Q60 0.38-0.40 0.54-0.57 0.34-0.39 Width 0.65-0.68 - 0.51-0.56 - 0.45-0.52 - Leg I Femur 1.52-1.75 1.34-1.53 1.22-1.35 1.05-1.25 1.14-1.31 1.02-1.13 Patella 1.96-2.16 1.60-1.86 1.62-1.70 1.31-1.55 1.27-1.55 1.14-1.30 Tibia 1.50-1.70 1.22-1.45 1.14-1.28 1.00-1.03 0.98-1.13 0.90-0.98 Tarsus-Basitarsus 1.12-1.27 1.01-1.12 0.82-1.03 0.82-0.98 0.93-0.97 0.82-0.87 Leg II Femur 1.02-1.17 0.97-1.10 0.72-0.92 0.80-0.84 0.79-0.88 0.73-0.84 Patella 0.59-0.67 0.52-0.61 0.36-0.50 0.45-0.52 0.44-0.53 0.45-0.52 Tibia 0.69-0.77 0.64-0.75 0.55-0.63 0.51-0.54 0.48-0.55 0.48-0.51 Basitarsus 0.58-0.69 0.56-0.65 0.46-0.53 0.42-0.52 0.45-0.52 0.40-0.48 Leg III Femur 0.95-1.05 0.89-1.03 0.70-0.74 0.65-0.79 0.70-0.78 0.66-0.76 Patella 0.41-0.53 0.40-0.46 0.28-0.36 0.35-0.38 0.34-0.38 0.32-0.40 Tibia 0.55-0.66 0.51-0.63 0.41-0.46 0.43-0.45 0.38-0.42 0.32-0.41 Basitarsus 0.64-0.77 0.62-0.71 0.46-0.57 0.46-0.55 0.45-0.54 0.43-0.51 Leg IV Femur 1.44-1.56 1.32-1.52 1.18-1.26 1.15-1.30 1.12-1.25 1.05-1.16 Patella 0.64-0.73 0.59-0.70 0.50-0.56 0.45-0.57 0.55-0.65 0.51-0.60 Tibia 1.02-1.15 0.98-1.09 0.83-0.95 0.81-0.95 0.77-0.85 0.73-0.80 Basitarsus 0.94-1.04 0.85-0.95 0.70-0.82 0.68-0.81 0.70-0.78 0.65-0.75 Comparisons.— This species is usually larger than other species in the group, with the exception of S. pentapeltis. Details of the male flagellum serve to distinguish it further from related forms. Also see under S. belkini. Distribution.— Known only from the type locality. Remarks.— Search of other palm oases in Joshua Tree National Monument has failed to produce schizomids, but if found they should closely resemble S. foshuensis. The type locality and possibly the only suitable habitat for this species is maintained by a fault- zone spring, which, if caused to go dry, would presumably eliminate this species. Lost Palm Oasis, another similar oasis in Joshua Tree National Monument, has been drained and made uninhabitable for this and perhaps other interesting relicts. This species has been collected in relatively low temperatures, where water was crystallized on the soil surface. In such cases it has been collected on the vertical face of partially exposed rocks. Additional records. -California'. San Bernardino County: Forty-nine Palms, Joshua Tree National Monument, 20 February 1970 (J. Rowland), two males, two females, one immature (MCZ), 22 February 1970 (J. Rowland, D. Harris), four females, two immatures (CAS), 30 December 1970 (J. Rowland), one female (TTU). Schizomus briggsi (Rowland) Figs. 15,18,21,30,36-37 Trithyreus belkini: Horn 1967: 216-220 [misidentification] . Trithyreus briggsi Rowland 1972b: 1-9; Rowland 1972c: 156; Dumitresco 1977: 151. Schizomus briggsi: Rowland and Reddell 1979a: 163. ROWLAND AND REDDELL-NEW WORLD SCHIZOMIDA IV 37 Figs. 34-40.-Parts of male schizomids of the briggsi group: 34-37, pedipalps: 34, 35, S. borregoensis : 34, right, lateral view: 35, right, dorsal view; 36, 37, S. briggsi : 36, right, lateral view; 37, right, mesal view of tibia and tarsus-basitarsus only; 38-40, lateral view of flagella: 38, S. pentapeltis; 39, S. belkini ; 40, S. joshuensis. 38 THE JOURNAL OF ARACHNOLOGY Table 7. -Measurements of the species of Schizomus not assigned to groups: 1, two males, S. infernalis ; 2, two females, S. infernalis ; 3, one female, Schizomus sp. from Sierra Nevada, Colombia; 4, two males, S. troglobius ; 5, three females, S. troglobius; 6, two males, S. armasi ; 7, three females, S'. arrmsi. Except as otherwise noted all measurements are of lengths. 1 2 3 4 5 6 7 Carapace 1.14-1.18 1.13-1.17 1.17 1.05-1.07 1.19-1.24 0.85-0.87 0.84-0.86 Flagellum Length 0.40-0.41 0.25-0.27 0.31 0.32-0.33 0.29-0.29 0.27-0.27 0.19-0.21 Width 0.23-0.23 - - 0.16-0.16 — 0.22-0.22 _ Leg I Femur 1.05-1.07 0.86-0.94 1.08 1.42-1.51 1.39-1.43 0.80-0.85 0.68-0.73 Patella 1.16-1.30 1.04-1.15 1.26 1.78-1.86 1.64-1.71 0.96-1.03 0.80-0.85 Tibia 0.95-0.95 0.78-0.82 0.93 1.31-1.35 1.27-1.29 0.72-0.75 0.58-0.62 Tarsus-Basitarsus 0.78-0.82 0.70-0.73 0.83 1.01-1.02 0.98-1.00 0.63-0.65 0.57-0.58 Leg II Femur 0.74-0.82 0.63-0.69 0.76 0.91-0.93 0.95-0.98 0.54-0.55 0.48-0.52 Patella 0.43-0.55 0.35-0.41 0.45 0.47-0.48 0.47-0.50 0.30-0.33 0.26-0.30 Tibia 0.44-0.50 0.41-0.43 0.44 0.63-0.63 0.62-0.65 0.32-0.35 0.29-0.31 Basitarsus 0.41-0.43 0.35-0.35 0.45 0.54-0.55 0.52-0.55 0.32-0.32 0.27-0.29 Leg III Femur 0.65-0.69 0.58-0.60 0.67 0.79-0.81 0.77-0.83 0.40-0.47 0.43-0.45 Patella 0.29-0.35 0.25-0.27 0.31 0.35-0.35 0.36-0.40 0.21-0.22 0.21-0.22 Tibia 0.27-0.35 0.29-0.30 0.34 0.48-0.49 0.51-0.53 0.25-0.27 0.22-0.24 Basitarsus 0.42-0.42 0.34-0.38 0.46 0.39-0.60 0.57-0.58 0.31-0.32 0.28-0.29 Leg IV Femur 1.04-1.04 0.92-0.95 1.11 1.30-1.32 1.32-1.34 0.72-0.85 0.70-0.76 Patella 0.51-0.54 0.41-0.46 0.50 0.51-0.53 0.52-0.57 0.34-0.35 0.33-0.35 Tibia 0.69-0.72 0.60-0.65 0.76 0.89-0.89 0.94-0.96 0.52-0.53 0.48-0.50 Basitarsus 0.60-0.65 0.50-0.59 0.67 0.80-0.82 0.80-0.83 0.46-0.61 0.41-0.43 Description. -Male. Color greenish brown. Carapace with three pairs of dorsal and three apical setae. Metapeltidium split. Eyespots indistinct. Anterior sternum with 13 bifid setae. Abdominal terga I- VII with two setae, terga VIII-IX with four setae, segment XII with acutely produced, well-developed posterodorsal process. Vestigial stigmata darker than sterna. Flagellum roughly hexagonal, with vague median depression followed by two elevations in line. Pedipalpal trochanter, femur, patella, and tibia extremely elongate; the tibia with a mesal spur apposible to tarsus-basitarsus; tarsal-basitarsal spurs about 1/6, claw about 1/4 length of tarsus-basitarsus. Tarsal-basitarsal segments of leg I of the following approximate proportions: 43-6-6-8-8-9-22. Other leg segment measurements given in Table 6. Female. Flagellum composed of four articles. Spermathecae composed of four pairs of small, broad lobes, each with minute terminal elevations, with no special sclerotization. Type data.— Holotype male and allotype female taken on the north face of Rocky Hill, 2.7 mi. E Exeter (1500 ft.), Tulare County, California, 28 January 1971 (P. and J. Rowland) (AMNH, examined). Comparisons.-^. briggsi is most similar to S. joshuensis and S. belkini, but is smaller in almost every respect than S. jsohuensis. It is easily distinguished from S. belkini by the abbreviation of the convex longitudinal ridge on the dorsal aspect of the male flagellum. In S. belkini the ridge is broad and extends from the base to the tip of the flagellum, whereas in S. briggsi the ridge is narrow and extends from near the middle of the ROWLAND AND REDDELL-NEW WORLD SCHIZOMIDA IV 39 flagellum to the tip. S . joshuensis also lacks a complete ridge, but differs from S. briggsi in having two lateral pits at the base of the ridge whereas S. briggsi has none. Distribution.— This species is known from near Academy, Fresno County, south to Fountain Springs, Tulare County, California, on scattered rock outcroppings along the west face of the Sierra Nevada foothills. Remarks.— briggsi is the northernmost New World schizomid. Its occurrence is seasonal, being abundant in winter and early spring, but apparently disappearing as the temperatures increase and humidities decrease. As with S. joshuensis , and perhaps other briggsi group species, S. briggsi has been found in relatively cold weather. K. Horn (pers. comm.) reported that he has collected this species under rocks with snow on the sur- rounding ground. Such temperature adaptations are not characteristic of schizomids in general. Additional records. -California: Fresno County: 7 mi. E. Academy, 16 April 1967 (T. Briggs), six adults (CAS); Squaw Valley, 23 March 1941 (S. Mulaik), four females (AMNH); 7 mi. N Piedra, 21 January 1967 (T. Briggs), one adult (TTU); 1.6 mi. SW Piedra, 21 January 1967 (T. Briggs), one adult (TTU), 21 January 1967 (T. Briggs, A. Jung, W. Lum, V. Lee, G. Leung, M. Wong, K. Horn), 21 adults (LACM); Tulare County: north face of Rocky Hill, 2.7 mi. E Exeter, 21 January 1971 (J. and P. Rowland), 21 adults (TTU), 28 January 1971 (J. and P. Rowland), 40 adults (LACM, AMNH, CAS), 5 January 1972 (J. and P. Rowland), five adults (TTU); north face of Rocky Hill, 2.1 mi. E town of Rocky Hill, 19 December 1966 (T. Briggs, V. Lee, K. Horn), 22 adults (MCZ); northwest face of Rocky Hill, 1.4 mi. E town of Rocky Hill, 22 January 1967 (T. Briggs, A. Jung, W. Lum, K. Horn), 18 adults (TTU); 12 mi. NE Hammond, 21 March 1941 (S. Mulaik), one adult (AMNH); 5 mi. NE Lemoncove, 20 March 1941 (S. Mulaik), three adults (AMNH); hill, 2 mi. SE Ivanhoe, 18 December 1966 (T. Briggs), two adults (CAS); 9 mi. N Woodlake, 22 March 1941 (S. Mulaik), three females (AMNH); hill, 3 mi. E Lindsay, 19 December 1966 (T. Briggs, V. Lee, K. Horn), ten adults (MCZ); 6.3 mi. E Fountain Spring, 19 March 1967 (T. Briggs), one adult (TTU); 7 mi. E Fountain Spring, 19 March 1967 (P. Lum, V. Lee, K. Horn), 24 adults (MCZ). Schizomus belkini (McDonald and Hogue) Figs. 15-16,32-33,39 Trithyreus belkini McDonald and Hogue 1957: 1-7,* Briggs and Horn 1966: 270, 273-274; Horn 1967: 216-220; Rowland 1971: 304, 308-309; Briggs and Horn 1972: 2; Rowland 1972b: 1, 4, 5, 7, 8; Rowland 1972c: 153, 155, 156, 158, 159. Schizomus belkini : Rowland and Reddell 1979a: 162. Description.— Male. Color brownish. Carapace with three pairs of dorsal and three apical setae. Eyespots indistinct. Anterior sternum with 13 bifid setae. Abdominal terga I-VII with two setae, terga VIII-IX with four setae, segment XII with acutely produced, well-developed posterodorsal process. Vestigial stigmata darker than sterna. Flagellum roughly hexagonal, with a pair of vague median depressions divided by a wide median ridge. Pedipalpal trochanter, femur, patella, and tibia extremely elongate; the tibia with a mesal spur apposible to tarsus-basitarsus; tarsal-basitarsal spurs about 1/6, claw about 1/4 length of tarsus-basitarsus. Tarsal-basitarsal segments of leg I of the following approxi- mate proportions: 42-5-7-7-7-8-20. Other leg segment measurements given in Table 6. Female. Flagellum composed of four articles. Spermathecae composed of four pairs of small, broad lobes, each with minute terminal elevations, no special sclerotization. Type data.— Hoi otype male, allotype female, and one male and three female paratypes taken at Crater Camp, Santa Monica Mountains, Los Angeles County, California, 21 March 1953, in oak humus (J. Belkin, R. Schick) (AMNH, examined). 40 THE JOURNAL OF ARACHNOLOGY Comparisons.-This species is generally the smallest of the members of the briggsi group. This species may be separated from S. joshuensis and S. briggsi , its closest relatives, by the details of the male flagellum. Distribution. -This species is known from the Santa Monica, Santa Ynez, and San Gabriel Mountains on mainland southern California, and on Santa Cruz Island off its coast. Remarks.— The widely disjunct populations indicate that this species has probably been restricting its range since xerothermic conditions have prevailed in southern California, except for a narrow moist corridor along the coastal foothills, which provides an avenue for northward dispersal. As in other members of this group, this species has only been found during winter and early spring months; it apparently retreats into subter- ranean habitats in warmer, drier weather. Additional records.- California: Los Angeles County: Santa Monica Mountains: 4.7 mi. N Topanga Beach, sycamore litter, 27 December 1966 (A. Jung, K. Horn), two males (MCZ), 7 April 1966 (T. Briggs, A. Jung, K. Horn), one male, one female, one immature (CAS); Topanga Canyon, 21 March 1953 (J. Belkin, R. Schick), one male, three females (AMNH), 19 December 1965 (T. Briggs, D. Owyang), one female (CAS), 29 March 1952 (R. Schick), one female, two immatures (AMNH), 27 February 1952 (J. Belkin, W. McDonald), two females, six immatures (AMNH); Malibu Canyon: Tapia Park, 4 April 1954 (L. Moskowski), one immature (AMNH), 7 March 1971 (J. Rowland, P. Brashier), two females, one immature (TTU), 16 February 1970 (J. Rowland, M. Brand), one male, one female, one immature (CAS); Santa Monica Mountains, April 1953 (R. Schick), one female, two immatures (AMNH); San Gabriel Mountains: Eaton Canyon, 28 February 1967 (M. Thompson), one immature (LACM), 30 March 1968 (J. Rowland, B. Firstman), two males, three females, one immature (AMNH); Santa Barbara County: Ose Canyon, Santa Ynez Mountains (=Oso Canyon, San Rafael Mountains), 26 December 1943 (W. S. Ross), one female (CAS); Santa Cruz Island, April 1913 (collector unknown), one male, one female (MCZ); Santa Cruz Island Field Station, 19 December 1967 (K. Horn), one male, two females, one immature (CAS); Raven’s Wood, Canada Del Puerto Canyon, 21 December 1967 (T. Briggs, A. Jung, K. Horn), one male, one female, one immature (CAS). UNPLACED SPECIES The following four taxa present considerable difficulties in their proper placement and are, therefore, not placed in any of the species groups discussed above. Schizomus troglobius, new species Figs. 41, 44, 48 Description.— Male. Color plae brownish. Carapace with three pairs of dorsal, the medians the smallest, and two apical setae. Eyespots absent. Anterior sternum with 11 bifid setae. Abdominal terga I-VII with two setae, VIII-IX with four setae, abdominal segments X-XII slightly elongated, segment XII without evidence of posterodorsal process. Vestigial stigmata darker than sterna. Flagellum laterally compressed, with complex sculpturing. Pedipalpal trochanter slightly produced distally; tarsal-basitarsal spurs about 1/5, claw about 1/3 length of tarsus-basitarsus. Tarsal-basitarsal segments of leg I of the following approximate proportions: 37-7-10-7-9-10-22. Other leg segment measurements given in Table 7. Female. Flagellum composed of three articles. Lateral spermathecae somewhat longer than medians, with sclerotized bulbs, median with unsclerotized bulbs, laterals curved inwardly. Type data.— Holotype male, allotype female, and six immatures taken in Jackson Bay Cave, Clarendon Parish, Jamaica, 21 or 22 December 1972 (S. and J. Peck) (AMNH); one ROWLAND AND REDDELL-NEW WORLD SCHIZOMIDA IV 41 adult male and one adult female paratypes with the same data (MCZ); one adult female paratype with the same data (TTU). Comparisons.— This species has a limited affinity to other Antillean species. It is the only Antillean species in which the female has a flagellum composed of three articles; the spermathecae, however, resemble those of other species in the dumitrescoae group in having the laterals longer than the medians. The lateral compression of the male flagellum and the slight elongation of pygidial segments are unknown in the other species from the Antilles. Distribution.— Known only from the type locality. Etymology.— The specific name is taken from the Greek troglo- meaning cave, and bios meaning life. Remarks.— This is the only apparent Jamaican troglobite. It may be more closely related to the mexicanus group than to the dumitrescoae group, as indicated by the flagella of both sexes. The male flagellum, however, is so highly derived that it gives no reliable clue. The slight elongation of the pygidial abdominal segments in the males adds further to the confusion of the proper placement of this species. Schizomus troglobius may, as with S. armasi, represent a relict of the mexicanus group which inhabited Carib- bean land masses before diversification of the dumitrescoae group. Schizomus infernalis Rowland Figs. 43, 46-47, 50 Schizomus infernalis Rowland 1975b: 18-20. Description.— Male. Color brownish. Carapace with two pairs of dorsal and two apical setae. Eyespots irregular, but circular. Anterior sternum with 13 bifid setae. Abdominal tergum I with two setae, II with four setae, III-VII with two setae, VIII-IX with four setae, segment XII without evidence of posterodorsal process. Vestigial stigmata darker than sterna. Flagellum spade shaped, with a pair of vague dorsal depressions. Pedipalpal trochanter very long, distinctly produced apically; femur greatly thickened, with one mesal and two lateral teeth; patella curved downward, expanded distally; tibia with mesal, subapical curved spur apposible to tarsus-basitarsus; tarsal-basitarsal spurs about 1/4, claw about 2/5 length of tarsus-basitarsus. Tarsal-basitarsal segments of leg I of the following approximate proportions: 42-8-9-9-9-10-19. Other leg segment measurements given in Table 7. Female. Flagellum composed of three articles. Median spermathecae just slightly longer than laterals, the former curved outwardly, the laterals with more distinct terminal bulbs, no special sclerotization. Type data.-Holotype male and allotype female taken 0.8 km N Ruinas de Palenque, near Palenque, Chiapas, Mexico, 25 July 1973, from Berlese samples (R. Mitchell, J. Reddell) (AMNH, examined); one male and three female paratypes with the same data (TTU, examined). Comparisons.— This species is unlike any other Mexican species in the massive develop- ment of the femur and trochanter of the male pedipalps and in the presence of four very strong dorsal setae on abdominal tergum II. The flagella of both males and females is similar to that of the species of the mexicanus group; the spermathecae, however, are different from mexicanus group species in having the median and lateral lobes nearly equal in size. 42 THE JOURNAL OF ARACHNOLOGY Figs. 41-47. -Parts of male schizomids: 41-46, flagella: 41-43, dorsal views: 41,5'. troglobius; 42, S. armasi-, 43, S. infernalis; 44-46, lateral views: 44, S. troglobius ; 45, S. armasi; 46, S. infernalis', 47, lateral view of right pedipalp of S. infernalis. ROWLAND AND REDDELL-NEW WORLD SCHIZOMIDA IV 43 Distribution.— Known only from the type locality. Remarks.— This species perhaps could be placed tentatively in the mexicanus group, but the highly derived condition of the male pedipalps, the dorsal setation of the abdomen, and the morphology of the female spermathecae obscure its true relationships. Schizomus sp. Fig. 51 Description.— Female. Color greenish. Carapace with three pairs of dorsal and two apical setae. Abdominal terga I- VII with two setae, terga VIII-IX with four setae. Vestigial stigmata lighter than sterna. Flagellum composed of four articles. Pedipalpal trochanter not produced dist ally; tarsal-basitarsal spurs about 1/5, claw about 2/5 length of tarsus- basitarsus. Tarsal-basitarsal segments of leg I of the following approximate proportions: 34-5-6-7-7-8-16. Other leg segment measurements given in Table 7. Spermathecae of aberrant form, perhaps with a small median highly convergent unsclerotized pair and a very large lateral highly sclerotized pair. Male unknown. Specimens examined.— One female and two immatures taken in the Sierra Nevada, Colombia (B. Malkin) (AMNH). Comparisons.— The greenish color, four articles of the flagellum, and spermathecal form serve to distinguish this species from other New World species. Distribution.— Known only from the Sierra Nevada, Colombia. Remarks.— The combination of characters which serve to distinguish this species so readily from other New World species also confuses its proper placement within existing groups. In the absence of males no attempt is made to place this species. Schizomus armasi, new species Figs. 42, 45, 49, 52 Description.— Male. Color brownish. Carapace with three pairs of dorsal and two apical setae. Metapeltidium split or entire. Eyespots distinct, round. Anterior sternum with nine bifid setae. Abdominal terga I-VLI with two setae, VIII-IX with four setae, segment XII without evidence of posterodorsal process. Vestigial stigmata darker than sterna. Flagellum spade shaped, with a pair of faint dorsal depressions with slight lateral elevations. Pedipalpal trochanter not produced distally; femur, patella, and tibia elongate; the tibia without spurs; tarsal-basitarsal spurs about 1/8, claw about 1/3 length of tarsus- basitarsus. Tarsal-basitarsal segments of leg I of the following approximate proportions: 25-5-6-6-5-6-13. Other leg segment measurements given in Table 7. Female. Flagellum composed of three articles. Median and lateral spermathecae joined basally, wide, not expanded distally, slightly divergent outwardly. Type data.— Holotype male and allotype female taken at Uvero, El Cobre, Oriente, Cuba, 25 May 1972 (L. de Armas) (IZC); one male, one female, and one immature paratypes with the same data (AMNH). Comparisons.— This species is closely related to two Cuban species not studied by us, S. rowlandi Dumitresco and S. orghidani Dumitresco. The excellent descriptions and figures given by Dumitresco (1973, 1977) provide clear evidence of their close relation- ship. The male flagellum of S. armasi is laterally angular and bears a pair of weak dorsal 44 THE JOURNAL OF ARACHNOLOGY 48 49 Figs. 48-52. -Parts of schizomids: 48-51, female spermathecae: 48,5'. troglobius; 49 , S. armasi', 50, S. infernalis\ 51, Schizomus sp. from Sierra Nevada, Colombia; 52, lateral view of male right pedipalp of S. armasi. ROWLAND AND REDDELL-NEW WORLD SCHIZOMIDA IV 45 depressions, whereas the flagellum of S. rowlandi is gradually curved laterally and apparently without dorsal relief. The flagellum of S. orghidani also lacks the angularity of S. armasi. The pedipalps of S. armasi are not produced distally, whereas they are in S. rowlandi and S. orghidani. All three species share a close similarity of the female spermathecae in that the median and lateral pairs are joined basally. S. armasi has three pair of dorsal carapacial setae, whereas S. rowlandi has only two pair. S. orghidani also has three pairs of dorsal carapacial setae, but the median pair are greatly reduced in size. The occasional split metapeltidium in S. armasi will also help distinguish this species from S. orghidani and S. rowlandi. Distribution.— Known only from the type locality. Etymology.— The specific name is a patronym given for Dr. Luis F. de Armas, the discoverer of this and other Cuban schizomids. Remarks. -This species should probably be placed with S. orghidani and S. rowlandi into a geographically and morphologically distinct group. This has not been done here because we have seen only S. armasi. These three species may represent ancient dichotomies from a proto -mexicanus group lineage, which may have inhabited Caribbean land masses and which may have given rise to a divergent line now largely extinct, having been replaced by dumitrescoae group species. ACKNOWLEDGMENTS We express our appreciation to Dr. Robert W. Mitchell for his assistance during the entire course of this study. The following curators made material available from their respective institutions: Dr. J. A. L. Cooke, American Museum of Natural History, New York, New York (AMNH); Dr. Ralph Crabill, National Museum of Natural History, Smithsonian Institution, Washington, D. C. (USNM); Dr. Luis F. de Armas, Instituto de Zoologfa, Academia de Ciencias, La Habana, Cuba (IZC); Dr. Willis J. Gertsch (AMNH); Dr. C. L. Hogue, Los Angeles County Museum, Los Angeles, California (LACM); Dr. Herbert W. Levi, Museum of Comparative Zoology, Harvard University, Cambridge, Massachusetts (MCZ); Dr. Robert W. Mitchell, The Museum, Texas Tech University, Lubbock, Texas (TTU); Dr. Norman I. Platnick (AMNH); and Dr. R. X. Schick, California Academy of Sciences, San Francisco, California (CAS). We are particularly grateful to the following individuals who contributed specimens from their private collections: T. S. Briggs, C. J. Goodnight, K. Horn, W. Icenogle, B. J. Kaston, W. Lum, and S. B. Peck. LITERATURE CITED Banks, N. 1900. Synopsis of North American invertebrates. American Nat., 39: 293-323. Briggs, T. S. and K. Horn. 1966. A new schizomid whip-scorpion from California with notes on the others (Uropygi: Schizomidae). Pan-Pacific Entomol., 42: 270-274. Briggs, T. S. and K. Horn. 1972. A carvernicolous whip-scorpion from the northern Mojave Desert, California (Schizomida: Schizomidae). Occas. Papers California Acad. Sci., No. 98, 7 p. Chamberlin, R. V. 1939. A new arachnid of the order Pedipalpida. Proc. Biol. Soc. Washington, 52: 123-124. Comstock, J. H. 1948. The spider book. 2nd ed., rev. by W. J. Gertsch. Ithaca, New York: Comstock. 729 p. Cook, O. F. 1899. Hubbardia, a new genus of Pedipalpi. Proc. Entomol. Soc. Washington, 4: 249-261. Dumitresco, M. 1977. Autres nouvelles especes du genre Schizomus des grottes de Cubia. Resultats des expeditions biospeologiques cubano-roumaines a Cuba, 2: 147-158. 46 THE JOURNAL OF ARACHNOLOGY Gertsch, W. J. 1940. Two new American whip-scorpions of the family Schizomidae. American Mus. Novitates, No. 1077, 4 p. Giltay, L. 1935. Notes arachnologiques Africaines. Bull. Mus. Hist. Nat. Belgique, 11(32): 1-8. Hansen, H. J. and W. Sorensen. 1905. The Tartarides, a tribe of the order Pedipalpi. Ark. Zool., 8: 1-78. Hilton, W. A. 1932. Tartarid whip-schorpions of southern California. J. Entomol. Zool. Claremont, 24: 33-34, 45-46. Horn, K. 1967. Notes on two California whip-scorpions. Pan-Pacific Entomol., 43: 216-220. Kishida, K. 1930. On the occurrence of the genus Trithyreus in Bonin Island. Lansania, Tokyo, 2: 17-19. (In Japanese) Kraus, O. 1957. Schizomidae aus Kolumbien (Arachnida, Pedipalpi-Schizopeltidae). Senck. Biol., 38: 245-250. McDonald, W. A. and C. L. Hogue. 1957. A new Trithyreus from southern California (Pedipalpida, Schizomidae). American Mus. Novitates, No. 1834, 7 p. Mello-Leitao, C. 1931. Pedipalpos do Brasil e algumas notas sobre a ordem. Arch. Mus. Nac., 33: 7-72. Moles, M. L. 1917. Another record of a small whip scorpion in California. J. Entomol. Zool. Clarement, 9: 1-7. Moles, M. L. 1921. A list of California Arachnida. II Pedipalpida. J. Entomol. Zool. Claremeont, 13: 11. Rowland, J. M. 1971. A new Trithyreus from a desert oasis in southern California (Arachnida: Schizomida: Schizomidae). Pan-Pacific Entomol., 47: 304-309. Rowland, J. M. 1972a. Brooding habits and early development of Trithyreus pentapeltis (Arachnida, Schizomida). Entomol. News, 83: 69-74. Rowland, J. M. 1972b. A new species of Schizomida (Arachnida) from California. Occas. Papers Mus. Texas Tech Univ., No. 5, 9 p. Rowland, J. M. 1972c. Origins and distribution of two species groups of Schizomida, (Arachnida). Southwestern Nat., 17: 153-160. Rowland, J. M. 1973a. A new genus and several new species of Mexican schizomids (Schizomida: Arachnida). Occas. Papers Mus. Texas Tech Univ., No. 1 1, 23 p. Rowland, J. M. 1973b. Revision of the Schizomida (Arachnida). J. New York Entomol. Soc., 80: 195-204. Rowland, J. M. 1973c. Three new Schizomida of the genus Schizomus from Mexican caves (Arachnida). Assoc. Mexican Cave Stud. Bull., 5: 135-140. Rowland, J. M. 1975a. Classification, phylogeny and zoogeography of the American arachnids of the order Schizomida. Ph.D. Dissertation. Lubbock: Texas Tech Univ., 415 p. Rowland, J. M. 1975b. A partial revision of Schizomida (Arachnida), with descriptions of new species, genus, and family. Occas. Papers Mus. Texas Tech Univ., No. 31, 21 p. Rowland, J. M. and J. R. Reddell. 1977. A review of the cavernicole Schizomida (Arachnida) of Mexico, Guatemala, and Belize. Assoc. Mexican Cave Stud. Bull., 6: 79-102. Rowland, J. M. and J. R. Reddell. 1979a. The order Schizomida (Arachnida) in the New World. I. Protoschizomidae and dumitrescoae group (Schizomidae: Schizomus) . J. Arachnol., 6: 161-196. Rowland, J. M. and J. R. Reddell. 1979b. The order Schizomida (Arachnida) in the New World. II. simonis and brasiliensis groups (Schizomidae: Schizomus). J. Arachnol., 7: 89-119. Rowland, J. M. and J. R. Reddell. 1980. The order Schizomida (Arachnida) in the New World. III. mexicanus and pecki groups (Schizomidae: Schizomus). J. Arachnol., 8:1-34. Takashima, H. 1943. Scorpionida and Pedipalpi of the Japanese Empire. Acta Arachnol., 8: 5-30. (In Japanese) Werner, F. 1935. Scorpiones, Pedipalpi, p. 1-490. In H. B. Bronns Klassen und Ordnungen des Tierreichs, bd. 5, abt. 4, buch 8, lief. 1-3. Leipzig: Akademische Verlagsgesellschaft. Manuscript received December 1978, revised November 1979. Muchmore, W. B. 1981. Cavernicolous species of Larca, Aracheloarca and Pseudogarypus with notes on the genera (Pseudoscorpionida, Garypidae and Pseudogarypidae). J. Arachnol., 9:47-60. CAVERNICOLOUS SPECIES OF LARCA , ARCHEOLARCA , AND PSEUDOGARYPUS WITH NOTES ON THE GENERA, (PSEUDOSCORPIONIDA, GARYPIDAE AND PSEUDOGARYPIDAE) William B. Muchmore Department of Biology University of Rochester Rochester, New York 14627 Abstract Six new species are described from caves in the western United States, as follows: Larca laceyi and Pseudogarypus orpheus from California, Archeolarca welbourni, A. cavicola, and Pseudogarypus hy- pogeus from Arizona, and Archeolarca guadalupensis from Texas. Larca granulata (Banks) is redes- cribed and other species of the genera are discussed. INTRODUCTION Over the past several years I have received for study several cavernicolous pseudoscorpions from western United States which deserve notice and description. Among these are representatives of Larca Chamberlin and Pseudograypus Ellingsen sent by Lawrence A. Lacey, of Larca sent by Andrew G. Grubs, and of Archeolarca Hoff and Clawson and Pseudograypus sent by W. Calvin Welbourn. In addition to describing the new forms I take this opportunity to report some observations made on these genera during the past 20 years. FAMILY GARYPIDAE HANSEN Subfamily Garypinae Simon Genus Larca Chamberlin For a recent review of this genus see Benedict and Malcolm 1977: 1 14-118. Larca chamherlini Benedict and Malcolm A single male referable to this species was found in Dirty Fissure, Calaveras County, California, 26 May 1977, by A.G. Grubbs. This is of the same size as the specimens from 48 THE JOURNAL OF ARACHNOLOGY Oregon and northern California described by Benedict and Malcolm (1977) but has somewhat more robust palps and legs. It is undoubtedly an epigean form, only accidental in the cave. It should be noted that this specimen has four setae in the flagellum on each chelicera, not three as mentioned by Benedict and Malcolm (1977:116) as characteristic of L. chamberlini. If the reduced number of setae is really characteristic of all the more north- ern forms, then this one may turn out to be a distinct species. Larca laceyi, new species Figs. 1-6 Materials.— Holotype male (WM 4652.01003) and eight paratypes (3 male, 4 female, 1 tritonymph) collected in Music Hall Cave, Calaveras County, California, 20 January 1973, by Lawrence A. Lacey. Types are deposited in the Florida State Collection of Arthro- pods, Gainesville. Diagnosis.— Generally similar to Larca granulata (Banks) and L. chamberlini Benedict and Malcolm with only two trichobothria on the movable chelal finger; but much larger than those species, with carapace more than 0.55 mm and palpal femur more than 0.85 mm in length. Description.— ADULTS: Males and females similar but females slightly larger. With the characters of the genus (see Benedict and Malcolm 1977:1 14). Well sclerotized and color- ed, palps reddish brown, other parts light brown. Carapace trapezoidal, with posterior width greater than length, surface covered with low granules and with a transverse furrow two-thirds distance from anterior margin; anterior margin nearly straight, slightly rough- ened near middle; with four strongly corneate eyes; about 40 prominent, curved setae, of which six are at anterior and 6-8 at posterior margin. Coxal area typical, widest across 4th coxae. Tergites 2-8 completely divided, 1 and 9 partly divided, 10 and 11 entire, surfaces covered with low granules; sternites 4-8 completely divided, others entire, surfaces reticu- lated to scaly; pleural membranes longitudinally plicate. Tergal chaetotaxy about 5:6: 10: 10:12:12:12:12:1 1 :T6T :9:2; setae prominent and curved as on carapace. Sternal chaetotaxy of male about 20: [3-3] :(0)20(0):(0)6(0):9:9:9:8:8:7:6:2; all setae small and delicate; setae on male genital opercula and internal genitalia as shown in Fig. 1 ; setae on female genital opercula as in Fig. 2; female internal genitalia marked by three well sclerotized cribriform plates in a transverse row and a smaller one behind (Fig. 2). Chelicera small, about 0.35 as long as carapace; four setae on hand, es quite long (two setae in position es on three specimens); fixed finger with two tiny subterminal denticles and four small teeth, movable finger with a subterminal roughened area; galea of female very long (about 2/3 as long as movable finger), straight, and with three small rami near tip; galea of male short and with variable serrations near tip; serrula exterior of about 15 blades; flagellum of four sparsely denticulate setae. Palp long and moderately slender (Fig.3);femur 1.52-1.59 and chela 1.65-1.74 times as long as carapace. Palpal trochanter 1.75-1.95, femur 4.85-5.5, tibia 3.55-3.95, and chela (without pedicel) 3.8-4.45 times as long as broad; hand (without pedicel) 2.35-2.8 times as long as deep; movable finger 0.89-0.94 as long as hand. Surfaces of palp strongly granulate except distal halves of fingers; setae mostly prominent, arcuate. Trichobothria as shown in Fig. 4; fixed finger with eight, movable finger with only two, presumbly sb and b. Fixed finger with 33-38 contiguous marginal teeth, retroconical distally and becoming flattened toward proximal end of row; movable finger with 35-38 similar teeth. Venom ducts in each finger inconspicuous, nodus ramosus 0.2 length of finger from tip. MUCHMORE-NEW CAVERNICOLOUS PSEUDOSCORPIONS 49 Legs long and slender (Figs. 5 and 6); leg IV with entire femur 3. 9-4. 5 and tibia 4. 7-5.1 times as long as deep. Proximal segments heavily granulate, becoming scaly distally. Setae of proximal segments large, arcuate, becoming smaller and straight distally; subterminal tarsal setae simple; no noticeable tactile setae present. Arolia twice as long as claws. TRITONYMPH: Much like the adults but paler, smaller, and with less attenuate appendages and fewer setae. Palpal femur 5.05, tibia 3.6, and chela (without pedicel) 4.3 times as long as broad. Movable chelal finger with two trichobothria as in the adult, fixed Finger with only seven, apparently ist missing. Chelicera with four setae on hand; flagellum of four setae; galea long and trifid at tip as in female. Measurements (mm).— Figures given first for the holotype male, followed in paren- theses by ranges for the adult paratypes. Body length 2.11(2.05-2.39). Carapace length 0.56(0.56-0.62). Chelicera 0.21(0.19-0.22) long. Palpal trochanter 0.33(0.33-0.355) by Figs. 1 - 6 -Larca laceyi, new species: 1. male genital opercula and internal genitalia; 2, female genital opercula; 3, dorsal view of right palp; 4, lateral view of left chela; 5, leg I; 6, leg IV. 50 THE JOURNAL OF ARACHNOLOGY 0.18(0.17-0.20); femur 0.89(0.865-0.96) by 0.17(0.16-0.19); tibia 0.71(0.69-0.76) by 0.19(0.185-0.21); fchela (without pedicel) 0.96(0.97-1.06) by 0.23(0.22-0.27); hand (without pedicel) 0.51(0.50-0.57) by 0.19(0.185-0.235); pedicel about 0.07 long; mov- able finger 0.47(0.47-0.525) long. Leg IV: entire femur 0.59(0.59-0.64) by 0.13(0.14-0.155); tibia 0.445(0.445-0.47) by 0.095(0.09-0.095). Tritonymph: Body length 1.65. Carapace length 0.52. Palpal femur 0.71 by 0.14; tibia 0.555 by 0.155; chela (without pedicel) 0.86 by 0.20; hand (without pedicel) 0.46 by ?; pedicel 0.06 long; movable finger 0.41 long. Etymology.— The species is named for Lawrence Lacey who collected all the speci- mens. Remarks. -Larca laceyi is definitely modified for cave life and is probably a troglobite. It may be compared with the local epigean form, L. chamberlini (see above and Benedict and Malcolm 1977): it is considerably larger than L. chamberlini ; it has six setae at the anterior margins of the carapace, rather than eight; the palps are longer with reference to the carapace; and the palpal and pedal segments are more slender. Like L. chamberlini , this species has only four setae on the cheliceral hand, which probably indicates a close relationship between the two. Also found in Music Hall Cave is a new species of Pseudogarypus, described below. Larca granulata (Banks) Gary pus granulatus Banks, 1 89 1 : 1 6 3. Larca granulata ; Chamberlin 1930:616, Hoff 1949:447, Hoff and Bolsterli 1956:163, Hoff 1958:15, Nelson 1975:282. The type locality for this species is Ithaca, Tompkins County, New York. It has also been reported from Illinois, Tennessee, and Michigan. New Records.— NEW YORK: Albany, Cattaraugus, Genesee, Monroe, Schuyler, and Wyoming counties. NEW HAMPSHIRE: Sullivan County. PENNSYLVANIA: Lycoming County. WEST VIRGINIA: Greenbrier, Mercer, and Pocohontas counties. VIRGINIA: Giles County. NORTH CAROLINA: Jackson, Macon, and Transylvania counties. KAN- SAS: Lincoln County. Inasmuch as no complete description has ever been published for this species, the type species of the genus, it seems appropriate to present such a description here. It is based mainly on the three syntypes (1 male, 2 females) in the Museum of Comparative Zoology at Harvard University, of which a female (WM1 2 16.01 001) is selected as the lectotype; also considered are a number of other specimens from upstate New York, within 100 miles from the type locality. Descriptions. -Males and females are quite similar except for a few sexually dimorphic characters and the slightly larger size of the latter. With general characters of the genus (see Benedict and Malcolm 1977:114). Well sclerotized and colored, palps light brown, other parts tan. Carapace trapezoidal with posterior width greater than length, surface heavily granulate and with a transverse furrow behind the middle; 4 strongly corneate eyes; 40 or more arcuate setae, usually 8 at anterior and 8 at posterior margins. Coxal area typical. Tergites 1-8 and sternites 4-8 at least partly divided; surfaces covered with low, irregular granulations; pleural membranes longitudinally plicate. Tergal chaetotaxy usually about 8:8:10:12:14:14:12:10:10:T6T:10;2, most setae rather heavy and arcuate; sternal chaetotaxy of female usually about 12:(0)14(0):(0)10(0):12:12:12:10:9:8:6:2, MUCHMORE-NEW CAVERNICOLOUS PSEUDOSCORPIONS 51 most setae fine and straight. Anterior gential operculum of male with irregular row of 12-20 small setae, genital opening internally with 1 or 2 setae on each side, posterior operculum with irregular row of 20-25 setae. Chelicera small, about 0.35 as long as carapace; 5 setae on hand, es quite long; fixed finger with a very narrow lamina exterior; fixed finger with 6-8 small teeth, movable finger with an irregular subapical lobe; galea slender and with 3 terminal rami, longer and better developed in female (Hoff 1949:448); serrula exterior of about 15 blades; flagel- lum of 4 setae, the distal one with several erect spinules along the free border. Palp long and moderately slender (Hoff 1949:448); femur 1.25-1.35 and chela 1.6- 1.7 times as long as carapace ; femur 4. 1-4. 6, tibia 2.9-3. 5, and chela (without pedicel) 3.64.2 times as long as broad; hand (without pedicel) 2. 1-2.4 times as long as deep; movable finger nearly as long as hand. All surfaces heavily granulate except distal parts of fingers; setae mostly arcuate. Trichobothria as shown in Hoff (1949:448); apparently, only sb and b present on movable finger. Each finger with 35-40 marginal teeth. Legs slender; leg I with basifemur about 1.25 times as long as telofemur; leg IV with entire femur 3. 6-4.0 and tibia 3.94.5 times as long as deep. Surfaces heavily granulate; most setae arcuate; no obvious tactile setae; subterminal tarsal setae simple ; arolia much longer than claws. Measurements (mm) of females.— Figures given first for the lectotype, followed in parentheses by ranges for eight others. Body length 2.03(1.91-2.12). Carapace length 0.57(0.51-0.59). Chelicera 0.205(0.19-0.215) long. Palpal trachanter 0.32(0.285-0.33) by 0.19(0.145-0.20); femur 0.75(0.655-0.775) by 0.165(0.145-0.175); tibia 0.61(0.52-0.65) by 0.185(0.16-0.20); chela (without pedicel) 0.91(0.86-0.96) by 0.25(0.22-0.265); hand (without pedicel) 0.465(0.445-0.50) by 0.21(0.205-0.245); pedicel about 0.06 long; mov- able finger 0.46(0.425-0.47) long. Leg I: basifemur 0.24(0.21-0.255) long; telofemur 0.19(0.17-0.19) long. Leg IV: entire femur 0.54(0.48-0.565) by 0.13(0.125-0.145); tibia 0.41(0.33-0.41) long. Males.— Ranges for five specimens. Body length 1.74-1.88. Carapace length 0.46-0.515. Chelicera 0.155-0.18 long. Palpal trochanter 0.26-0.29 by 0.14-0.165; femur 0.605-0.67 by 0.14-0.15; tibia 0.49-0.54 by 0.155-0.175; chela (without pedicel) 0.78-0.87 by 0.18-0.21 ; hand (without pedicel) Q40-0.445 by 0.175-0.19; pedicel about 0.05 long. Leg I: basifemur 0.19-0.22 long; telofemur 0.155-0.175 long. Leg IV: entire femur 0.425-0.475 by 0.1 1-0.125; tibia 0.32-0.355 long. Remarks.— It is impossible, on the basis of published accounts, to distinguish clearly between L. granulata and L. lata (Hansen) from Europe (see Beier 1963). Probably only a direct comparision of representatives of the two species will reveal any differences which may exist. Most specimens of Larca granulata have been found in dry situations under rocks or logs, or in debris in old stumps or logs, occasionally in association with mice or chip- munks. Larca notha Hoff This species has until recently been known only from a single specimen (male) from Larimer County, Colorado (Hoff 1961). Mention (but no description) of a single adult from Harney County, Oregon, is made by Benedict (1978). New record.— CANADA: Saskatchewan, Val Marie, 10 June 1955, 3 males 2 females, 3 tritonymphs, and 1 deutonymph collected by J.R. Vockeroth from the nest of bank swallows [Canadian National Collection of Insects] . 52 THE JOURNAL OF ARACHNOLOGY Diagnosis.— The only obvious difference between this species and L. granulata is in the number of trichobothria on the movable finger of the chela, three in the former, two in the latter. Supplemental description. -The description of the holotype male by Hoff is very complete and nothing need be added except to note that reexamination of the specimen reveals five setae on the hand of the chelicera, as in L. granulata. The males from Saskatchewan are quite similar in all respects to the holotype. The females show some sexual differences and are a little larger than males. Chaeto- taxy of genital opercula and anterior sternites 10:(0)12(0):(0)4(0):-. Cheliceral hand with five setae; galea long and trifid terminally. The nymphs are much like the adults but smaller and with reduced numbers of trichobothria on the chelal fingers. Tritonymphs are lacking both t and st from the movable finger and ist from the fixed finger. All nymphs have five setae on the hand of the chelicera and four setae in the flagellum. Measurements of adults from Saskatchewan (mm). -Body length 1.85-1.95. Carapace length 0.465-0.51. Palpal femur 0.59-0.63 by 0.16-0.175; tibia 0.50-0.54 by 0.185-0.19; chela (without pedicel) 0.80-0.83 by 0.235-0.24; hand (without pedicel) 0.415 by 0.21-0.215; pedicel 0.055 long; movable finger 0.415-0.42 long. Leg IV: entire femur 0.46 by 0.12-0.125. Genus Archeolarca Hoff and Clawson For a recent review of this genus see Benedict and Malcolm (1977:1 18-1 19). Archeolarca rotunda Hoff and Clawson This species has been reported from Utah County, Utah, Bernalillo County, New Mexico, and Deschutes County, Oregon. In addition it can be mentioned here that several specimens were collected by G.F. Knowlton from pack rat nests in shallow caves in Blacksmith Fork Canyon, Cache County, Utah, in June 1970. These are somewhat varied in morphology but are certainly referable to this species. Archeolarca welbourni, new species Figs. 7 and 8 Material. -Holotype female (WM 4453.01003) and six paratype females found on wall in Malmquist Fissure, 25 September 1975 and 31 January 1976; one paratype male from “lower level, dark zone” of Lomacki Fissure, 27 September 1975; three paratype females from Dangling Flake Crack, 21 January 1976; one male and two female paratypes from Sipapu Cavern, 31 January 1976 -- all locations in Wupatki National Monument, Coconino County, Arizona, and all collections made by W. Calvin Welbourn. The types are in the Florida State Collection of Arthropods, Gainsville. Diagnosis.— Similar to A. rotunda but significantly larger, with palpal femur greater than 0.9 mm in length. Description.— With the characters of the genus (Benedict and Malcolm 1977:118). Males and females similar but females larger. Generally well sclerotized and colored, palps and carapace reddish brown, other parts light brown. Carapace trapezoidal, with posterior width greater than length; anterior margin slightly concave; surface heavily MUCHMORE-NEW CAVERNICOLOUS PSEUDOSCORPIONS 53 granulate and with a distinct transverse furrow 0.6 the distance from anterior margin; with four bulging eyes; with 25-30 setae, usually including six at anterior and four at posterior margin. Coxal area typical, widest across 4th coxae. Abdomen typical; tergal chaetotaxy of female holotype 5:7:8: 11:12:10: 13:1 1 : 1 2 :T4T : 7:2; sternal chaetotaxy 16:(0)9(0): (0)4(0):8:7:8:9:8:6:2:2; others are varied but similar. Internal genitalia marked by four cribriform plates, as in Lorca (see Fig. 2), but posterior smaller plate appearing fragmented. Chaetotaxy of anterior sternites of male about 24: [3-3]: (0)20(0):(0)6(0):8:8 -; internal genitalia much like that in Larca (see Fig. 1). Figs. 7 and 8.- Archeolarca welbourni, new species: 7, dorsal view of left palp; 8, lateral view of right chela. Figs. 9 and 10 -Archeolarca guadalupensis , new species: 9 dorsal view of right palp; 10, lateral view of left chela. Figs. 11 and 12.— Archeolarca cavicola, new species: 11, dorsal view of right palp; 12, lateral view of left chela. 54 THE JOURNAL OF ARACHNOLOGY Chelicera small, about 0.35 as long as carapace; four setae on hand, es quite long: galea very long, slender, with three small rami near tip; flagellum of four setae, of which the distal one or two are sparsely denticulate; narrow lamina exterior present; serrula exterior of 17-18 blades. Palp long and moderately slender (Fig. 7), femur about 1.45 and chela about 1.75 times as long as carapace; femur 4. 3-4.9, tibia 3. 2-3. 8, and chela (without pedicel) 3.3-3 .9 times as long as broad; hand (without pedicel) 1.8-2.25 times as long as deep; movable finger about 0.95 as long as hand. Surfaces strongly granulate, except for chelal fingers; setae mostly prominent, arcuate. Trichobothria as in Fig. 8. Fixed finger with about 35 contiguous, cusped marginal teeth; movable finger with about 30 similar teeth and 2-3 low, rounded ones proximally. Venom ducts present in each finger but inconspicuous, nodus ramosus about 0.1 5 length of finger from tip. Legs long and slender; leg IV with entire femur about 5.0 and tibia about 6.0 times as long as deep. Surfaces mostly scaly; setae arcuate, rather conspicuous; no noticeable tactile setae; subterminal tarsal setae simple ; arolia twice as long as claws. Measurements (mm) of females.— Figures given first for the holotype, followed in parentheses by ranges for 10 paratypes. Body length 2.70(2.58-2.89). Carapace length 0.70(0.68-0.74). Chelicera 0.26(0.25-0.27) by 0.13(0.12-0.13). Palpal femur 1.03 (0.925-1.07) by 0.21(0.20-0.25); tibia 0.88(0.78-0.90) by 0.23(0.23-0.28); chela (with- out pedicel) 1.21(1.20-1.31) by 0.32(0.31-0.39); hand (without pedicel) 0.62(0.63-0.70) by 0.29(0.29-0.37); pedicel about 0.09 long; movable finger 0.61(0.58-0.65) long. Leg I: basifemur 0.37(0.35-0.40) long; telofemur 0.26(0.24-0.29) long. Leg IV: entire femur 0.75(0.71-0.83) by 0.15(0.15-0.16); tibia 0.61(0.55-0.63) by 0.10(0.09-0.11). Males. -Body length 2.32-2.58. Carapace length 0.615-0.65. Chelicera 0.215-0.235 by 0.11-0.12. Palpal femur 0.90-0.985 by 0.19-0.21; tibia 0.76-0.84 by 0.21-0.25; chela (without pedicel) 1.10-1.175 by 0.27-0.32; hand (without pedicel) 0.57-0.63 by 0.25-0.28; pedicel about 0.08 long; movable finger 0.54-0.605 long. Leg I: basifemur 0.34-0.385 long; telofemur 0.235-0.27. Leg IV: entire femur 0.68-0.78 by 0.13-0.15; tibia 0.52-0.55 by 0.09-0.105. Etymology.— The species is named welbourni in honor of W. Calvin Welbourn who has collected many pseudoscorpions of importance in the southwestern United States. Remarks.— This is the second species to be described in the genus Archeolarca. As A. rotunda is regularly found in pack rat nests, it is not unexpected to find the new species in a cave. It is easy to conceive that pack rats have carried representatives of the genus down into the caves, where isolation and speciation have occurred. It is not known whether A. welbourni continues to associate with pack rats in the cave, as the specimens were found on bare walls. Archeolarca guadalupensis, new species Figs. 9 and 10 Material.— Holotype female (WM 4602.02004) and six paratypes (2 male, 1 female, 2 tritonymph) from Lower Sloth Cave, Guadalupe Mountains National Park, Culberson County, Texas, 17 April 1976 (W.C. Welbourn). The types are in the Florida State Collec- tion of Arthropods, Gainesville. Diagnosis.— Similar to A. welbourni but slightly smaller (palpal femur 0.81-0.96 mm long) and with more slender palpal segments (chela l/w=4 .0-4.7). MUCHMORE-NEW CAVERNICOLOUS PSEUDOSCORPIONS 55 Description.— With the characters of the genus (Benedict and Malcolm 1977:118). Males and females similar but females larger and more robust. Generally well sclerotized and colored, palps and carapace reddish brown, other parts light brown. Carapace trape- zoidal, with posterior width greater than length; anterior margin slightly concave, surface heavily granulate and with a distinct transverse furrow 0.6 the distance from anterior margin; with four corneate eyes, about equal in size; with 30-35 setae, six at anterior and 4-6 at posterior margin. Coxal area typical. Tergal chaetotaxy of holotype female 4:6:8:9:11:10:11 :9:8:T3T:8:2; sternal chaetotaxy 12:(0)8(0):(0)4(0):7:6:7:5:9:6:3:2. Cribriform plates of internal genitalia as in A. welbourni. Chaetotaxy of anterior sternites of male about 20: [3-3] :(0)14(0):(0)4(0):5:7:-; internal genitalia like that of A. wel- bourni. Chelicera small, about 0.37 as long as carapace; four setae on hand; galea of female long, slender, and trifid at tip, that of male short and unequally bifid; flagellum of four setae, the distal one sparsely denticulate; serrula exterior of 15-16 blades. Palp long and slender (Fig. 9), femur about 1 .4 and chela about 1 .65 times as long as carapace; femur 4.75-5.05, tibia 3.45-3.85 and chela (without pedicel) 4.0-4. 7 times as long as broad; hand (without pedicel) 2.7-3.15 times as long as deep; movable finger about 0.8 as long as hand. In lateral view especially, base of chelal hand tapering, not sharply set off from pedicel (Figs. 9 and 10); surfaces lightly granulate, except for chelal fingers; setae thin, mostly arcuate. Trichobothria as in Fig. 10. Fixed chelal finger with about 30 cusped marginal teeth, movable finger with about 25 similar teeth and 3-4 low, rounded ones proximally. Venom ducts inconspicuous. Legs long and slender: leg IV with entire femur 4.6-5 .0 and tibia 5. 5-5.9 times as long as deep. Arolia twice as long as claws. Measurements (mm) of females.— Figures given first for the holotype, followed in parentheses by those for the paratype. Body length 2.71(2.51). Carapace length 0.67(0.69). Chelicera 0.25(0.25) by 0.13(0.13). Palpal femur 0.90(0.96) by 0.19(0.20); tibia 0.76(0.81) by 0.22(0.21); chela (without pedicel) 1.08(1.12) by 0.27(0.28); hand without pedicel) 0.67(0.70) by 0.25(0.25); pedicel about 0.075 long; movable finger 0.53(0.55) long. Leg I: basifemur 0.33(0.34) long; telofemur 0.235(0.245) long. Leg IV: entire femur 0.70(0.73) by 0.14(0.16); tibia 0.53(0.54) by 0.09(0.10). Male.— Body length 2.23-2.44. Carapace length 0.60-0.65. Chelicera 0.22-0.245 by 0.12. Palpal femur 0.71-0.95 by 0.16-0.19; tibia 0.67-0.78 by 0.18-0.21 ; chela (without pedicel) 0.98-1.10 by 0.21-0.24; hand (without pedicel) 0.60-0.69 by 0.20-0.22; pedicel about 0.075 long; movable finger 0.48-0.53. Leg I: Basifemur 0.295-0.35 long; telofemur 0.20-0.24 long. Leg IV: entire femur 0.60-0.72 by 0.12-0.15; tibia 0.46-0.56 by 0.08-0.10. Etymology. -The species is named guadalupensis for the Guadalupe Mountains, where it is found. Remarks.— Though it is generally smaller than the other two known cavernicolous species of Archeolarca, A. guadalupensis has the most attenuated palpal segments; in this respect it is more modified than the others as a cave dwelling form. Archeolarca awco/a,new species Figs. 11 and 12 Material.— Holotype female (WM 5398.01001) from Cave of the Domes, Grand Can- yon National Park, Coconino County, Arizona, 15 April 1978 (W. Calvin Welbourn). The type is in the Florida State Collection of Arthropods, Gainesville. 56 THE JOURNAL OF ARACHNOLOGY Diagnosis. -Similar to A. welbourni but larger (palpal femur 1 .09 mm in length) and hand of palpal chela more rounded, especially at base. Description of female (male unknown). -Generally well sclerotized and colored, palps and carapace reddish brown, other parts light brown. Carapace trapezoidal; anterior mar- gin nearly straight; surface heavily granulate and with a shallow transverse furrow 0.6 the distance from anterior margin; with four corneate eyes, but posterior pair much smaller than anterior; about 26 vestitural setae, with six at anterior and four at posterior margin. Coxal area typical of genus. Abdomen typical: tergal chaetotaxy 4:7:8:11:10:10:11:10: 10:T5T: 10:2; sternal chaetotaxy 1 5:(0)10(0):(0)4(0):6:7:7:7:6:6:2:3. Cribriform plates of internal genitalia as in A. welbourni. Chelicera typical, about 0.35 as long as carapace; four setae on hand; galea very long and slender, with two small, subterminal rami; flagellum of four setae; serrula exterior of 21 blades. Palp long and moderately slender (Fig. 11); femur 1.45 and chela 1.8 times as long as carapace; femur 4.65, tibia 3.45 and chela (without pedicel) 3.45 times as long as broad; hand (without pedicel) 1.95 times as long as deep; movable finger 1.01 times as long as hand. In lateral view especially, base of chelal hand gently rounded, not sharply set off from pedicel (Figs. 11 and 12). Surfaces strongly granulate, except for chelal fingers; setae arcuate, conspicuous. Trichobothria as in Fig. 12. Fixed chelal finger with 37 and movable finger with 35 cusped marginal teeth. Venom ducts inconspicuous. Legs long and slender; leg IV with entire femur 5.4 and tibia 7.0 times as long as deep, arolia more than twice as long as claws. Measurements (mm).-Body length 3.03. Carapace length 0.755. Chelicera 0.27 by 0.14. Palpal trochanter 0.455 by 0.265; femur 1.09 by 0.235; tibia 0.925 by 0.27; chela (without pedicel) 1.37 by 0.40; hand (without pedicel) 0.72 by 0.37; pedicel 0.11 long; movable finger 0.73 long. Leg I: basifemur 0.46 long; telofemur 0.295 long. Leg IV: entire femur 0.86 by 0.16; tibia 0.665 by 0.095. Etymology. -The species is named cavicola in recognition of its subterranean habitat. Remarks.— Of the three cavernicolous species of Archeolarca here described, A. cavi- cola shows the greatest overall adaption to the special habitat. Compared to either A. welbourni or A. guadalupensis , it is larger, has longer appendages, has more reduced posterior eyes, and has fewer setae on the carpace. FAMILY PSEUDOGARYPIDAE CHAMBERLIN Genus Pseudogarypus Ellingsen For a recent review of this family and genus, see Benedict and Malcolm (1978). Pseudogarypus orpheus, new species Figs. 13 and 14 Material.— Holotype male (WM 4657.01001) and paratype female from Music Hall Cave, 5 miles E of Parrots Ferry, Calaveras County, California, 24 December 1976 and 20 January 1973 respectively (Lawrence A. Lacey). The specimens are deposited in the Florida State Collection of Arthropods, Gainesville. MUCHMORE-NEW CAVERNICOLOUS PSEUDOSCORPIONS 57 Diagnosis. -Much like P. bicornis (Banks), but larger and more attenuated than most specimens of that species, with palpal femur 1 .55 mm long and 6.0 times as long as broad. Description.— With the general characters of the genus (Benedict and Malcolm 1978) and the following notable features. Male and female similar. Carapace about 1 .25 times as long as posterior breath; anterior margin with relatively deep notch between anterolateral and median protuberances; eyes well developed. Abdomen 1.2 times as long as broad; pleural membranes of male showing indistinct sclerites anteriorly, those of both sexes with numerous, scattered, tiny, thickened plaques. Male with anterior genital operculum bearing about 55 scattered, small setae, half of them concentrated near the middle of the hind margin; five long setae on each internal, crescent-shaped sclerite; posterior opercu- lum bearing about 50 scattered setae, some concentrated at the middle of the front margin. Female anterior operculum with about 50 small setae scattered over entire sur- face, and posterior operculum also with about 50 scattered setae. Chelicera typical (Morris 1948: Figs. 16-19); 0.43 as long as carapace; hand with 1 1 fine setae; flagellum of two slender, curved setae; galea simple, gently curved, longer in female than in male. Palp somewhat attenuate (Fig. 13), femur 1.95 and chela (with pedicel) 2.3 times as long as carapace; trochanter 1.45, femur 6.0, tibia 2.9, and chela (with pedicel) 5.0 times as long as broad; hand (with pedicel) 2.3 times as long as deep; movable finger 1.6 times as long as hand. Fixed chelal finger with 46 and movable finger with 36 spaced, generally conical marginal teeth; each finger with a large, terminal “venedens.” provided with a conspic- uous modified seta alongside (called a “lamina defensor” by Chamberlin 1931 : 133), but without an obvious venom duct. Trichobothria positioned as is usual in the genus (Fig. 14), including a close-set pair of short, accessory trichobothria on external surface of fixed finger near distal end. Each coxa I with five or six spines. Legs moderately attenu- ate; leg IV with basifemur 3.2, telofemur 3.6, tibia 6.7 and tarsus 14.0 times as long as deep. Measurements (mm).— Figures given first for holotype male, followed in parentheses by those for paratype female. Body length 3.07(2.98). Carapace length 0.87(0.76); pos- terior breadth 0.63(0.58). Chelicera 0.33(0.325) by 0.155(0.16). Palpal trochanter 0.40(0.40) by 0.28(0.27); femur 1.51(1.47) by 0.25(0.25); tibia 0.74(0.695) by 0.25(0.25); chela (without pedicel) 1.71(1.70) by 0.36(0.355); hand (without pedicel) 0.665(0.66) by 0.33(0.325); pedicel 0.08 long; movable finger 1.05(1.065) long. Leg IV: basifemur 0.42(0.41) by 0.13(0.13); telofemur 0.58(0.59) by 0.16(0.16); tibia 0.70(0.70) by 0.105(0.105); tarsus 0.89(0.93) by 0.065(0.065). Etymology. -The species is named for Orpheus, the Greek musician who went to the underworld in search of his wife. Remarks.-Attention should be called to the occurrence near the tip of the fixed chelal finger of a pair of short, accessory trichobothria. These were first reported by Hoff (1946) in P. bicornis and then by Morris (1948) for Neopseudogarypus scutellatus. Curi- ously, Benedict and Malcolm (1978) make no mention of these setae, although they figure them on the palps of all species treated. Such accessory trichobothria are found elsewhere among the pseudoscorpions only in members of the Chthonioidea. It is inter- esting to speculate that this common occurrence indicates some close phylogenetic relationship between the pseudogarypids and the chthonioids. Such speculations are strengthened by the occurrence in the two groups of distinct spines on the coxae of the anterior legs and by the similar appearance of internal guard setae in the male genitalia. Also of interest, but of unknown significance, is the occurrence of tiny thickened plaques in the pleural membranes of this species (these are distinct from the larger, 58 THE JOURNAL OF ARACHNOLOGY heavier, sclerotized plates which are seen in the male). Again it is tempting to speculate that these plaques represent the vestiges of setae, such as may be found in the pleural membranes of Gary pus calif ornicus (see Lee 1979) and Anagarypus oceanusindicus Chamberlin (unpublished observation). The chelicerae of pseudogarypids are not “very small” as stated by Benedict and Malcolm (1978:87), but actually are of fair size, being a third or more as long as the carapace. In the intact animal they do, however, appear very small because they are mostly hidden under the front edge of the carapace. Though it is true that venom ducts and glands are not apparent in pseudogarypids, there is some reason to believe that a functional venom apparatus may be present. Each of the chelal fingers has a long, sharp terminal “fang”, within which can be seen faint lines suggestive of a duct. That the duct cannot be followed into the finger may be due simply to the heavy, rough cuticle which covers the surface. It should be noted too that each “fang” is accompanied by a modified seta of the kind called “lamina defensor” by Chamberlin (1931 : 133). In other pseudoscorpions this structure is typically developed in association with a functional venedens; where the venom apparatus is reduced or absent, Figs. 13 and 14 -Pseudogarypus orpheus, new species: 13, dorsal view of right palp; 14, lateral view of left chela. Figs. 15 and 16 .-Pseudogarypus hypogeus, new species: 15, dorsal view of right palp; 16, lateral view of left chela. MUCHMORE-NEW CAVERNICOLOUS PSEUDOSCORPIONS 59 also the “lamina defensor” is reduced or absent. Though the function of this modified seta has never been demonstrated, I suggest that it might act as a trigger for the release of venom. It is strategically placed so that it is moved in a particular way when the venedens penetrates the body of a prey animal; nerve impulses from its base might then stimulate the release of venom through the venedens. If this is so, then presence of a well developed “lamina defensor” should indicate presence of a functional venom apparatus. In addition to this species and Larca laceyi (see above), the caves of Calaveras County, California, are home to several other troglobitic pseudoscorpions, namely Neochthonius troglodytes Muchmore (1969a), Microcreagris grahami Muchmore (1969b) and unde- scribed species of Apochthonius and Aphrastochthonius. Pseudogarypus hypogeus, new species Figs. 15 and 16 Material.— Holotype female (WM 4452.01001) and tritonymph from Doney Fissure, 27 September 1975, and four paratypes (1 male, 2 females, 1 tritonymph) from Dangling Flake Crack, 31 January and 2 October 1976 -- both locations in the Wupatki National Monument, Coconino County, Arizona and all collections by W. C. Welbourn. The types are in the Florida State Collection of Arthropods, Gainesville. Diagnosis.— With longer and more attenuated appendages than the local epigean form of P. bicornis\ length of palpal femur about 1 .3 mm, 1/w about 5.2, and femur about 1 .9 times as long as carapace. Description.— With the general characters of the genus (Benedict and Malcolm 1978:85) and the following noteworthy features. Male and female similar but female a little larger. Carapace about 1.25 times as long as posterior width; anterior margin with rather deep notch between anterolateral and median protuberances; eyes well developed. Abdomen typical; pleural membranes without obvious sclerites, but with occasional, tiny, thickened plaques. Anterior genital operculum with about 60 and posterior operculum with about 50 scattered, small setae. Chelicera 0.43 as long as carapace; hand with 10 setae; flagellum of two curved setae; galea simple, gently curved. Palp as in Fig. 15; femur about 1.9 and chela about 2.25 times as long as carapace; trochanter 1.4-1.45, femur 5.1-5.55, tibia 2.4-2.65, and chela (with pedicel) 4. 7-5.0 times as long as broad; hand (with pedicel) 2.1-2.15 times as long as deep; movable finger about 1.7 times as long as hand. Fixed chelal finger with 41-47 and movable finger with 32-37 weakly spaced, marginal teeth; each finger with a large “venedens” accompanied by a conspicuous, modified seta (“lamina defensor”). Trichobothria placed as usual, including the accessory pair on the fixed finger (Fig. 16). Legs typical; each coxa I with five or six spines; leg IV moderately attenuate with basifemur 2.95-3.2 and telofemur 3.4-3.55 times as long as deep. Measurements (mm). -Figures given first for the holotype, followed in parentheses by ranges for the three adult paratypes. Body length 2.88 (2.62-2.99). Carapace length 0.74(0.69-0.74). Chelicera 0.32(0.295-0.32) by 0.16(0.155-0.16). Palpal femur 1.39 (1.24-1.33) by 0.25(0.24-0.25); tibia 0.66(0.59-0.63) by 0.25(0.23-0.25); chela (with- out pedicel) 1.59(1.44-1.57) by 0.33(0.32-0.35); hand (without pedicel) 0.58(0.56- 0.58) by 0.315(0.295-0.31); pedicel about 0.075 long; movable finger 1.02(0.91-1.0) long. Leg IV: basifemur 0.40(0.37-0.40) by 0.125(0.125); telofemur 0.55(0.50-0.54) by 0.155(0.15-0.155). 60 THE JOURNAL OF ARACHNOLOGY Etymology.— The species is named hypogeus in recognition of its habitat beneath the surface, in earth cracks. Remarks.-The new species has noticeably longer and more attenuate appendages than representatives of a nearby population of the epigeanP. bicornu (Banks). In the latter the palpal femur is about 1.1 mm long, 4.85 times as long as broad, and 1 .55 times as long as carapace; and the chela (with pedicel) is about 1.4 mm long, 4.4 times as long as broad, and 1 .95 times as long as carapace. ACKNOWLEDGMENTS I am greatly indebted to Andrew G. Grubbs, Lawrence A. Lacey and W. Calvin Wel- bourn for sending me pseudoscorpions from western United States caves. Credit for the illustrations belongs to Charlotte H. Alteri. LITERATURE CITED Banks, N. 1891. Notes on North American Cher netidae. Canadian Entomol., 23:161-166. Beier, M. 1963. Ordnung Pseudoscorpionidea. Bestimmungsbucher zur Bodenfauna Europas, 1:1-313. Benedict, E.M. 1978. A biogeographical study of currently identified Oregon pseudoscorpions with an emphasis on western Oregon forms. Ph.D. Dissertation, Portland State University, pp.i-xv and 1-143. Benedict, E.M. and D.R. Malcolm. 1977. Some garypoid false scorpions from western North America (Pseudoscorpionida: Garypidae and Olpiidae). J. Arachnol., 5:113-132. Benedict, E.M. and D.R. Malcolm. 1978. The family Pseudogarypidae (Pseudoscorpionida) in North America with comments on the genus Neopseudogarypus Morris from Tasmania. J. Arachnol., 6:81-104. Chamberlin, J.C. 1930. A synoptic classification of the false scorpions or chela-spinners, with a report on a cosmopolitan collection of the same - Part II. The Diplosphyronida (Arachnida- Chelonethida). Ann. Mag. Nat. Hist. (ser. 10), 5:1-48 and 585-620. Chamberlin, J.C. 1931. The arachnid order Chelonethida. Stanford Univ. Publ. Biol. Sci. 7, no. 1:1-284. Hoff, C.C. 1946. A study of the type collections of some pseudoscorpions originally described by Nathan Banks. J. Washington Acad. Sci., 36:195-205. Hoff, C.C. 1949. The pseudo scorpions of Illinois. Bull. Illinois Nat. Hist. Surv., 24:407-498. Hoff, C.C. 1958. List of the pseudoscorpions of North America north of Mexico. Amer. Mus. Novit- ates, 1975:1-50. Hoff, C.C. 1961. Pseudoscorpions from Colorado. Bull. Amer. Mus. Nat. Hist., 122:409-464. Hoff, C.C. and J.E. Bolsterli. 1956. Pseudoscorpions of the Mississippi River drainage basin area. Trans. Amer. Micros. Soc., 75:155-179. Lee, V.F. 1979. The maritime pseudoscorpions of Baja California, Mexico (Arachnida: Pseudoscor- pionida). Occ. Pap. California Acad. Sci., 131:1-38. Morris, J.H.C. 1948. A new genus of pseudogarypin pseudoscorpions possessing pleural plates. Pap. Proc. Roy. Soc. Tasmania, 1947:43-47. Muchmore, W.B. 1969a. The pseudo scorpion genus Neochthonius Chamberlin (Arachnida, Chelone- thida, Chthoniidae) with description of a cavernicolous species. Amer. Midi. Nat., 81:387-394. Muchmore, W.B. 1969b. New species and records of cavernicolous pseudo scorpions of of the genus Microcreagris (Arachnida, Chelonethida, Neobisiidae, Ideobisiinae). Amer. Mus. Novitates, 2392:1-21. Nelson, S.O., Jr. 1975. A systematic study of Michigan Pseudoscorpionida (Arachnida). Amer. Midi. Nat., 93:257-301. Manuscript received September 1979, revised January 1980. Galiano, M. E. 1981. Revision del genero Phiale C. L. Koch, 1846 (Araneae, Salticidae) III. Las especies polimorficas del grupo mimica. J. Arachnol., 9:61-85. REVISION DEL GENERO PHIALE C. L. KOCH, 1846 (ARANEAE, SALTICIDAE) HI. LAS ESPECIES POLIMORFICAS DEL GRUPO MIMICA Maria Elena Galiano1 Museo Argentino de Ciencias Naturales “Bernardino Rivadavia” Av. Angel Gallardo 470, 1405 Buenos Aires, R. Argentina ABSTRACT Polymorphism is documented in species of the genus Phiale. Controlled rearing of Phiale tristis Mello-Leitao, 1945 demonstrated the existence of a chromatic sex-controlled polymorphism, with polymorphic females and monomorphic males. There is strong evidence that other three species are also polymorphic: Phiale mimica (Koch), P. crocea Koch and Phiale ortrudae, new species. These four species together with Phiale bulbosa (Cambridge), new combination (of which only the male holotypus is known), form a homogeneous group within the genus Phiale. Seven specific names are synonymized: Phiale mutilloides Mello-Leitao, 1947 = P. tristis Mello-Leitao, 194 5;P. rubrosericea Mello-Leitao, 1947 = P. tristis; P. nigrosigillata Mello-Leitao, 1947 = P. tristis; Freya bella (C. L. Koch, 1846) = P. mimica (C. L. Koch, 1846); Freya obscura (Taczanowski, 1872) = P. crocea C. L. Koch. 1846; Chira luctuosa Simon, 1902 = P. crocea; P. zonata Caporiacco, 1947 = P. crocea. EXTRACTO Se documenta el polimorfismo en especies de Salticidae del genero Phiale. La cria controlada de Phiale tristis Mello-Leitao, 1945, demostro la existencia de polimorfismo cromatico controlado por el sexo, con hembras polimorficas y machos monomorficos. Se evidencia que otras tres especies son tambien polimorficas: Phiale mimica (Koch), Phiale crocea Koch y Phiale ortrudae nueva especie. Estas cuatro especies, junto con Phiale bulbosa (Cambridge), nueva combination (de la cual solo se conoce el holotipo macho), forman un grupo homogeneo dentro del genero Phiale. Se establece la sinonimia de siete nombres especificos: Phiale mutilloides Mello-Leitao, 1947 = P. tristis Mello-Leitao, 1945; P. rubrosericea Mello-Leitao, 1947 = P. tristis; P. nigrosigillata Mello-Leitao, 1947 = P. tristis; Freya bella (C. L. Koch, 1846) = P. mimica (C. L. Koch, 184 6), Freya obscura (Taczanowski, 1872)= P. crocea C. L. Koch, 1846; Chira luctuosa Simon, 1902 = P. crocea; P. zonata Caporiacco, 1947 = P. crocea. 1 Miembro de la carrera del Investigador del Consejo Nacional de Investigaciones Cientificas y Tecnicas. Buenos Aires. 62 THE JOURNAL OF ARACHNOLOGY INTRODUC C ION En la primera contribution de esta serie (Galiano 1978) se demostro que Phiale mimica (C. L. Koch, 1846) y Phiale gratiosa C. L. Koch, 1846, durante anos consideradas como sinonimos, son entidades diferentes y facilmente distinguibles por la estructura de los epiginos y espermatecas. Ejemplares con el dibujo abdominal de P. mimica han sido hallados en colecciones de Panama, Venezuela, Colombia, Guyana, Guyana Francesa, Brasil, Bolivia, Paraguay y Argentina. Todos los especimenes son femeninos y nunca se encontraron machos que pudieran asignarse a la especie. Era logico pensar que los machos pudieran haber sido descriptos con otros nombres, por ser lo suficientemente diferentes como para no ser reconocidos. Con el objeto de elucidar este problema, se planeo una investigation con los siguientes pasos: (1) viajes a la selva subtropical misionera, donde se hallan ejemplares con el diseno de P. mimica aunque con epigino algo distinto; (2) captura de hembras fecundadas; y (3) cria de la descendencia. El plan se cumplio en todas sus etapas y los resultados superaron las expectativas. Se obtuvieron machos que, segun lo supuesto, eran muy diferentes a las hembras y habian sido descriptos como Phiale tristis Mello-Leitao, 1945, especie algo dudosa de la cual no se conocian las hembras (ver comentarios de pag. 76 )• Mas interesante aun fue com- probar que entre las hembras criadas, algunas presentaban un colorido totalmente distinto al de la madre: eran uniformemente negras, con una cubierta de pelitos negros y blancos mezclados en mechoncitos, sin las bandas amarillas caracteristicas de P. mimica . En un segundo viaje se capturaron todos los especimenes que a simple vista se reconocieron como Phiale y en el laboratorio se crio la descendencia de todos los desoves. Esta experiencia en mayor escala demostro la existencia en P. tristis de polimorfismo controlado por el sexo, con machos monomorficos y hembras polimorficas. Las variantes responden a patrones fijos de diseno y colorido, algunos de ellos exactamente iguales a los de especies de Phiale descriptas por autores anterioxes para otras areas geograficas. Asi, una de las variantes es igual a P. crocea C. L. Koch, 1846; tres responden a los colores de P. mutilloides Mello-Leitao, 1947, P. nigrosigillata Mello-Leitao, 1947 y P. rubrosericea Mello-Leitao, 1947, respectivamente; otra es similar a P. gratiosa (C. L. Koch, 1846); de todas estas especies solo se conocen las hembras. Por otro lado, los machos de P. tristis presentan el mismo patron de colorido que Frey a bulhosa (Cambridge, 1901), Freya bella (C. L. Koch, 1846) y Freya obscura (Taczanowski, 1872) de las cuales solo se conocen los machos. Frente a estos hallazgos se realizo una nueva busqueda en las colecciones de material conservado y se agruparon los especimenes de ambos sexos, con patrones de colorido semejantes a los observados en las crias deP. tristis. Pudo verse que se estaba en presencia de un grupo de taxa estrechamente emparentados, de distribution alopatrida, con polimorfismo cromatico controlado por el sexo. Habiendo comprobado que el color no es un caracter que permita separar las especies de este grupo, se buscaron otros caracteres y se hallo que la estructura de los genitales (pese a su gran uniformidad) presenta diferencias que son especificas y que concuerdan con el area de distribution. En los epiginos la diferenciacion es menor, pero los palpos de los machos permiten distinguir cinco entidades que se comportan como alopatridas con respecto a las demas del grupo, con muy escasa superposition (Fig. 9). Las entidades diferenciadas por la estructura del palpo del macho y el area de distri- bution son las siguientes: Phiale mimica (C. L. Koch, 1846), GALIANO-LAS ESPECIES POLIMORFICAS DE PHIALE 63 Phiale crocea (C. L. Koch, 1846), Phiale tristis Mello-Leitao, 1945, Phiale bulbosa (F. O. P. Cambridge, 1901), nueva combination, Phiale ortrudae nueva especie. El problema sistematico que plantea el estudio de estos taxa es el nivel de diferen- ciacion alcanzado. ^Se trata de verdaderas especies reproductivamente aisladas o mas bien es una especie politipica con cinco subespecies ? Es imposible averiguar si el grado de aislamiento reproductive ha llegado al nivel especifico o si puede haber cruzamientos en las areas de contacto. No existe ningun procedimiento valido para obtener una respuesta. La extension del polimorfismo a cuatro de las especies (de P. bulbosa se conoce solamente el tipo macho), la similitud entra las formas o variantes presentes en cada una de ellas, la semejanza de los machos monomorficos, la uniformidad en la estructura de los genitales y la distribution alopatrida podrian justificar el empleo del nivel subespecifico. Sin embar- go, las diferencias en la forma de los palpos son del mismo grado que las que habitual- mente se usan en Salticidae para separar las especies y la ausencia de formas intermedias (con exception de los ejemplares de Bolivia), parecen indicar que estas entidades son especies diferenciadas y ese es el criterio adoptado en este trabajo. Esta position podra por cierto ser apoyada o rechazada por medio de otras experiencias de cria y cruza- miento, que se lleven a cabo con todas las especies. Para su mejor comprerision, el trabajo se ha dividido en dos partes; la primera trata el polimorfismo y la segunda, la taxonomia del grupo de especies. PRIMERA PARTE: ESTUDIO DEL POLIMORFISMO Material y metodos.— Las aranas empleadas en esta experiencia fueron colectadas en el Parque Nacional Iguazu, provincia de Misiones, R. Argentina, en los meses de octubre de 1977 y 1978. La especie ha sido determinada como Phiale tristis Mello-Leitao, 1945 (ver comentarios sobre esta especie). Algunos individuos fueron colectados inmaduros y real- izaron la ultima muda en el laboratorio, donde se hicieron ensayos de copula y se obtuvieron desoves, pero los huevos utilizados en esta experiencia provinieron exclusiva- mente de hembras que se capturaron ya fecundadas y que desovaron en cautiverio. Las hembras adultas fueron colocadas en frascos de vidrio de medio litro de capacidad, tapados con tela de voile. En el interior, un trozo de papel arrugado proporciono punt os de apoyo para la construction de los refugios. La alimentation consistio en Musca domes- tica y la humedad se suministro por medio de un trozo de algodon embebido en agua. La iluminacion fue la natural en una habitation soleada y la temperatura ambiente se mantuvo por encima de los 20°C en invierno, llegando en ocasiones en el verano a los 35°C. Producida la dispersion de las crias, se re tiro a la madre del frasco. Las crias se dejaron todas juntas hasta despues de la segunda o tercera muda a contar de la dispersion, pues se comprobo que el numero de sobrevivientes era mayor por este sistema que cuando se las aislaba en tubos individuales desde el principio de su vida independiente. La alimen- tacion se hizo con Drosophila melanogaster en los primeros estadios y con Musca domestica en los ultimos. La mortalidad de los juveniles es muy grande. La experiencia se comenzo con diecisiete desoves, pero solamente se conseguieron adultos de nueve de ellos. Los juveniles son uniformemente negros, con los palpos blancos, los metatarsos I y II y todos los tarsos amarillos. Aproximadamente en la mitad de su desarrollo, empiezan a esbozarse en las hembras las bandas o manchas que caracterizaran su patron de diseno. En cada muda se producen pequenas modificaciones pero no son de importancia. Los machos 64 THE JOURNAL OF ARACHNOLOGY subadultos se reconocen por los palpos dilatados; su color es negro, cubierto por pelitos negros en el opistosoma y negros y blancos en el prosoma. Los dibujos de las figuras 1 a 8 son esquematicos; puede haber pequenas modifica- ciones en la forma y dimension de las manchas o bandas. El colorido deriva del color del tegumento mas los pelos que lo cubren; puede observarse correctamente solo en los primeros dias despues de la exuviation, ya que los pelos se desprenden facilmente. En los ejemplares vivos, el dorso de prosoma y opistosoma se pela por el habito de abrirse camino por pasajes estrechos, empujando frontalmente hacia adelante y arriba. Los animales preservados pierden los pelos con los anos o por fallas en la fijacion y ademas, los colores se alteran con el alcohol. El tegumento es negro en los especimenes vivos y se vuelve pardo en alcohol. En el prosoma pueden existir areas mas claras, en forma de bandas laterales o como una herradura detras de la region cefalica. En el opistosoma pueden haber bandas o manchas donde el tegumento es translucido (en alcohol blanquecino o amarillento). Los pelos son del tipo “plumoso” con brillo de seda, de colores negro, pardo olivaceo, bianco, amarillo, anaranjado y rojo. Los pelos negros se implantan siempre sobre el tegumento negro. Los pelos de color pueden estar sobre areas negras o claras, en cuyo caso el colorido resultante es diferente. Por ejemplo las cuatro manchas del opistosoma P-1 (Fig. 8) son de color rojo brillante, porque los pelos rojos se implantan sobre tegumento amarillo. En cambio, en el opistosoma “0”, el dorso es negro cubierto por pelos rojos y el efecto final es rojo oscuro, apagado. En los machos el diseno es producto de bandas o manchas de pelos blancos, amarillentos o pardo olivaceo, sobre el tegumento uniformemente negro. Resultados.— Para cada experiencia se menciona el ejemplar madre con su numero de colecta, el total de crias del desove y el de los especimenes que llegaron a la madurez. Los machos de todas las experiencias son iguales y se describen en la parte correspondiente a Phiale tristis. Experiencia N? 1 (Fig. 1)-Madre 642: Tegumento negro cubierto por pelos negros; tegumento de las bandas y manchas amarillo, con pelos amarillos. Pata I anaranjada; femur negro; patella parda. Crias: total 23. Machos: 4. Hembras: 642-1, como la madre (2 ejemplares). Hembras: 642-2, prosoma con tegumento negro, cubierto por pelos negros y blancos, mezclados en mechoncitos; predominio de pelos negros en la region cefalica y de pelos blancos en el margen. En el declive toracico, una banda longitudinal de pelos blancos. Opistosoma con tegumento negro, con pelos negros y blancos mezclados en mechoncitos. Pata I anaranjada, femur y extremo de la tibia, negros. Experiencia 2 (Fig. 2)-Madre N? 709: tegumento negro, cubierto por pelos negros y blancos, mezclados en mechoncitos. Pata I negra, metatarsos y tarsos amarillos. Crias: total 31. Machos: 5. Hembras: 709-1, prosoma con tegumento negro, el dorso densamente cubierto por pelos blancos. Opistosoma con tegumento negro con pelos negros; banda ancha de tegumento amarillo cubierta por pelos blancos. Pata I negra; metatarso y tarso amarillos, con apices pardos (6 ejemplares). Hembra: 709-2, prosoma con tegumento negro; region cefalica y bandas laterales de pelos amarillos. Opistosoma, tegumento negro con pelos negros; bandas amarillas con pelos amarillos. Pata I pardo claro; femur negro (2 ejemplares inmaduros). Figs. 1-4 .-Phiale tristis: Resultado de las experiencias de cria N- 1 a 4. Ver explication en el texto. GALIANO-LAS ESPECIES POLIMORFICAS DE PHIALE 65 66 THE JOURNAL OF ARACHNOLOGY Experiencia 3 (Fig. 3)-Madre 778: prosoma con tegumento negro cubierto por pelos rojos. Opistosoma negro, dorso con pelos rojos, lados con pelos negros y un par de manchas apicales de pelos blancos. Pata I negra; metatarso y tarso pardo claro. Crias: total 11. Machos: ninguno. Hembras: 778-1, tegumento negro, cubierto por pelos negros, blancos y rojos, mezclados en mechoncitos. Pata I negra; metatarso y tarso pardo claro (3 ejemplares). Hembras: 778-2, como la madre (3 ejemplares). Hembras: 778-3, prosoma con tegumento negro, region cefalica con pelos rojos; bandas marginales de pelos blancos. Pata I negra; metatarso y tarso pardo claro. Opistosoma como la madre (2 ejemplares). Experiencia 4 (Fig. 4)— Madre 789: tegumento negro, cubierto por pelos negros y blancos, mezclados en mechoncitos. Pata I negra; metatarso y tarso amarillos, con apices pardos. Crias: total 26. Machos: 10. Hembras: 789-1, prosoma con tegumento negro, cubierto por pelos negros y amarillos, mezclados en mechoncitos; banda toracica media y bandas marginales de pelos amarillos. Opistosoma, tegumento negro con pelos negros y amarillos, mezclados. Los pelos amarillos se concentran hacia el apice, formando grandes manchas laterales y cuatro manchas dorsales en forma de triangulo. Pata I parda; tibia negra (3 ejemplares). Hembras: 789-2 prosoma con tegumento negro; region cefalica cubierta por pelos rojos, bandas marginales de pelos blancos. Opistosoma con tegumento negro con pelos negros; bandas amarillas con pelos rojos. Pata I pardo oscuro; metatarso y tarso pardo claro (1 ejemplar). Hembras: 789-3, prosoma con tegumento negro cubierto por pelos negros y blancos, mezclados en mechoncitos; banda toracica media y bandas laterales de pelos blancos. Opistosoma con tegumento negro, con pelos negros y blancos mezclados en mechoncitos y dos pares de manchas laterales subapicales de pelos blancos. Pata I negra; metatarso y tarso pardo claro (2 ejemplares). Hembras: 789-4, prosoma con tegumento negro, con pelos negros y blancos mezclados en mechoncitos. Opistosoma y patas como el anterior (3 ejemplares). Experiencia Pft 5 (Fig. 5)-Madre N? 649: prosoma con tegumento negro cubierto por pelos rojos. Opistosoma con tegumento negro, cubierto por pelos negros; banda basal amarilla, cubierta por pelos anaranjados. Pata I anaranjada; tibia negra. Crias: total 33. Machos: 12. Hembras: 649-1 (a), tegumento negro, con pelos negros y amarillos mezclados en mechoncitos. Pata I negra; metatarso y tarso pardo claro (4 ejemplares). (b), el mismo diseno, pero pelos blancos en lugar de amarillos (1 ejemplar). Hembras: 649-2, prosoma con tegumento negro, cubierto por pelos rojos. Opistosoma con tegumento negro con pelos negros; banda amarilla con pelos anaranjados (1 ejem- plar). Hembras: 649-3, prosoma con tegumento negro con pelos negros; region cefalica con pelos rojos. Opistosoma como el anterior (1 ejemplar). Hembras: 6494, prosoma con tegumento negro cubierto por pelos rojos; opistosoma como la madre (5 ejemplares). Hembras: 649-5, opistosoma con tegumento negro con pelos negros; banda amarilla con pelos anaranjados. Pata I anaranjada; tibia negra (2 ejemplares). Experiencia N? 6 (Fig. 6 ) -Madre N? 712: tegumento negro, con pelos negros y blancos mezclados en mechoncitos. Pata I negra; metatarso y tarso pardo claro. Crias: total 40. Machos: 6. Hembras: 712-1, como la madre (13 ejemplares). Hembras: 712-2, prosoma con tegumento negro cubierto por pelos negros; banda toracica media, Figs. 5-7 .-Phiale tristis: Resultado de las experiencias de cria N- 5 a 7. Ver explicacion en el texto. GALIANO-LAS ESPECIES POLIMORFICAS DE PHIALE 67 68 THE JOURNAL OF ARACHNOLOGY bandas marginales y manchas entre los ojos laterales, con pelos blancos. Opistosoma, tegumento negro con pelos negros; manchas blanquecinas con pelos blancos. Pata I negra; metatarso y tarso pardo claro (2 ejemplares). Hembras: 712-3, prosoma con tegumento negro, con pelos negros y escasos pelos blancos, mezclados en mechoncitos. Pata I anaran- jada; femures negros (1 ejemplar). Experiencia 7 (Fig. 7)-Madre 779: tegumento negro, con pelos negros y blancos, mezclados en mechoncitos. Pata I negra; metatarso y tarso amarillos. Crias: total 17. Machos: 5. Hembras: 779-1, tegumento negro, con pelos negros, blancos y escasos rojos, mezclados en mechoncitos; en el apice del opistosoma, dos pares de manchas de pelos blancos. Pata I pardo claro; femur negro. (2 ejemplares). Hembras: 779-2, prosoma con tegumento negro, cubierto por pelos negros, blancos y rojos mez- clados; bandas marginales de pelos blancos. Opistosoma con tegumento negro, con pelos negros, blancos y rojos mezclados; apicalmente dos pares de manchas de pelos blancos salpicadas con escasos pelos rojos (2 ejemplares). Hembras: 779-3, prosoma como el anterior. Opistosoma con tegumento negro cubierto de pelos negros, blancos y rojos, y ademas con escasos pelos amarillos y anaranjados, todos mezclados. En los costados, apicalmente, dos pares de manchas y en el dorso una banda y dos triangulos de pelos blancos salpicados con escasos rojos. Pata I parda; femur negro; metatarso y tarso amaril- los (2 ejemplares). Experiencia N? 8/-Madre TV? Ill : tegumento negro cubierto por pelos negros y blancos mezclados en mechoncitos. Crias: total 31. Machos: 7. Hembras: 777-1, como la madre (2 ejemplares). Hembras: 777-2, tegumento negro, cubierto por pelos negros y amarillos, mezclados en mechoncitos (2 ejemplares). Hembras: 777-3, como 712-3 (2 ejemplares inmaduros). Experiencia 9 -Madre Eft 795: tegumento negro, con pelos negros y bancos, mez- clados en mechoncitos. Crias: total 10. Machos: 1. Hembras: ninguna. Discusion.— Los resultados obtenidos concuerdan con la definition de polimorfismo: presencia de dos o mas formas o variantes sincronicas y simpatridas, que son discontmuas (es decir, no pasan gradualmente de unas a otras) y en tal proportion que la mas rara de ellas no puede ser mantenida solamente por mutation recurrente (Ford 1940, 1945, 1953). En el caso de Phiale tristis, los machos son monomorficos y las hembras polimorficas, lo que evidencia un polimorfismo controlado por el sexo. El recuento de las variantes observadas demuestra que la mas comun es la de tegu- mento negro cubierto por pelos negros y blancos mezclados. Todos los ejemplares adultos comprendidos en esta experiencia (madres y crias) fueron medidos, su quetotaxia anotada, los epiginos aclarados y dibujados. Este estudio demostro que las diferencias morfologicas son minimas, las proporciones de las partes son constantes y las estructuras genitales tienen leves modificaciones individuales (Figs. 25-27). Por lo tanto, el polimorfismo es solamente cromatico. El polimorfismo en otras especies de Phiale del grupo mimica—E\ estudio de colecciones que abarcan amplias areas de America, asi como de los ejemplares tipicos de especies antiguas, no mencionadas desde su description, permitio delimitar cinco especies estrechamente relacionadas, en todas las cuales esta presente el polimorfismo cromatico controlado por el sexo. La gran variedad de formas encontradas surge en parte de las posibles combinaciones entre los disenos y colores de los prosomas con las de los opisto- somas. En la figura 8 y su leyenda se ilustran y describen las variantes polimorficas GALIANO-LAS ESPECIES POLIMORFICAS DE PHIALE 69 observadas; en la description de cada especie se enumeran las formas halladas. Muchas de las variantes son comunes a las cuatro especies (de P. bulbosa no se conocen las hembras) y aunque no se han hecho experiencias de cria con P. crocea, P. mimica y P. ortrudae, creo que no hay dudas sobre la correction del criterio empleado. En ejemplares conser- vados, se han localizado veintidos variantes en P. crocea , doce enP. mimica y cuatro en P. ortrudae (se dispuso unicamente del lote tipico). La distribution de las especies se indica en la figura 9. El material estudiado debe considerarse escaso para arribar a una con- clusion sobre el tipo de polimorfismo involucrado pero es suficiente como para demostrar su existencia en el grupo. Nada puede adelantarse sobre su significado y su posible valor adaptativo o selectivo. Toda conclusion en ese sentido requeriria una nueva linea de experimentacioa El polimorfismo en especies de otros generos y familias.—Existen en la literatura menciones sobre variantes en el patron de diseno y colorido de diversas especies de aranas, pero pocas veces hay referencias concretas o pruebas de que la causa sea el polimorfismo. En Phidippus audax (Salticidae) han sido notadas diferencias en el colorido debido a diferente distribution de las manchas y al color de los pelos escamosos (Taylor y Peck 1975, Hill 1978). Los ejemplares del sur son mas vivamente coloreados que los del norte, pero no ha podido discernirse un patron definido. Parecidas variaciones han sido descriptas para diversas especies de Araneidae: Eustala anastera y E. cepina parecen presentar tres “formas” de colorido (Levi Larinia directa tambien tiene tres coloraciones posibles (Levi 197 5a)‘, Araneus illaudatus muestra tan grandes diferencias de tamano entre individuos de un mismo sexo que Levi (1975b) se pregunta si puede tratarse de un polimorfismo geneticamente determinado para la pro- portion de crecimiento. En Gasteracantha versicolor y en Isoxya tabulata (Araneidae) Emerit (1969, 1973) ha hallado marcadas variaciones en el colorido, debido a la diferente extension de las areas melanicas. Platnick y Shadab (1975) suponen que Gnaphosa fontinalis (Gnaphosidae) es una especie polimorfica, ya que hay variaciones en los genitales de ambos sexos que se corresponden con diferentes patrones de colorido. En Steatoda (Theridiidae) Levi (1959) ha hecho un estudio de las variaciones en los genitales de machos y hembras y en los disenos del opistosoma. Parece haber varias especies involucradas y dines geograficos para ciertos caracteres. Enoplognatha ovata (Theridiidae) segun Levi (1957) presenta gran variation individual en palpos, epiginos y coloration. Ejemplares de esta especie fueron criados por Seligy (1969, 1971) quien comprobo que hay tres variantes de color que aparecen en la descen- dencia de un mismo cocon. Segun Tweedie (1970) las proporciones en que estas formas se presentan, demuestran que el caracter que las controla depende de dos genes alelomorfos. En una Oxyopidae, Oxyopes papuanus y una Zodariidae, Stores annulipes Kolosvary (1932) hallo un elevado numero de variaciones en el colorido del opistosoma. En Pisauridae, Blandin (1977) menciona la existencia de polimorfismo en Chiasmopes lineatus y Rothus purpurissatus. La primera de estas especies presenta por lo menos tres tipos de ornamentation, presentes en ejemplares de la misma localidad. R. purpurissatus tiene variaciones en el patron de colorido de ambos sexos, que ha hecho que seis formas fueran descriptas como especies. Afropisaura valida muestra tambien gran variation en la ornamentation, con dos casos extremos, vinculados por una serie de formas intermedias (Blandin 1976). 70 THE JOURNAL OF ARACHNOLOGY Fig. 8.-Variantes polimorficas de especies dePhiale del grupo mimica : A, macho, prosoma negro con manchas de pelos blancos. Opistosoma negro, cubierto dorsalmente por pelos pardos o pardo -olivaceos; bandas laterales y banda dorsal de pelos blancos amarillentos. B-P, Opistosomas de hembras. B y C, tegumento negro con pelos negros; bandas con tegumento y pelos amarillos. D y E, tegumento negro con pelos negros; bandas con tegumento amarillo cubierto de pelos anaranjados. F, tegumento negro con pelos negros; banda amarilla con pelos anaranjados. G-l, tegumento negro con pelos negros; banda amarilla con pelos blancos. G-2, tegumento negro con pelos negros; banda amarilla con pelos anaranjados. G-3, tegumento negro con pelos negros; banda amarilla con pelos amarillos. H, tegumento negro con pelos negros; banda y pelos amarillos. I, tegumento negro con pelos negros; banda amarilla con pelos anaranjados. J, tegumento negro con banda amarilla, todo cubierto por pelos rojos. K, tegumento negro con pelos negros; bandas amarillas con pelos rojos. L, tegumento negro con pelos negros; bandas amarillas con pelos anaranjados. M-l, tegumento negro, con pelos negros y blancos, mezclados en mechoncitos. M-2, tegumento negro, con pelos negros y amarillos, mezclados. M-3, tegumento negro, con pelos negros, blancos y rojos, mezclados. M-4, tegumento negro, con pelos negros, blancos, amarillos, anaranjados y rojos, mezclados. N, tegumento negro con pelos negros; bandas y manchas blanquecinas con pelos blancos. O, tegumento negro, el dorso cubierto por pelos rojos que a veces cubren tambien los costados. Otros ejemplares con pelos negros y dos pares de manchas apicales de pelos blancos en los costados. P-1, tegumento negro con pelos negros; manchas amarillas con pelos rojos. P-2, tegumento negro con pelos negros; manchas amarillas con pelos anaranjados. P-3, tegumento negro con pelos negros; manchas amarillas con pelos amarillos. 1-20, Prosomas de hembras. 1, tegumento negro con pelos negros; bandas laterales y media pardo claro con pelos amarillos. 2, tegumento negro con pelos negros; banda media pardo claro con pelos amarillos. 3, tegumento negro, totalmente cubierto por pelos blancos. 4, tegumento negro cubierto por pelos anaranjados. 5, tegumento negro, cubierto por pelos rojos. 6, tegumento negro, cubierto por pelos amarillos. 7, tegumento negro; region cefalica con pelos anaranjados; region toracica con pelos negros. 8, tegumento negro; r. cefalica con pelos amarillos; r. toracica con pelos negros. 9, tegumento negro; r. cefalica con pelos rojos; r. toracica con pelos negros. 10, tegumento negro; r. cefalica con pelos blancos; r. toracica con pelos negros. 11, tegumento negro; r. cefalica con pelos blancos; r. toracica con pelos negros y blancos, mezclados. 12, tegumento pardo o negro; r. cefalica con pelos rojos; bandas marginales con pelos blancos. 13, tegumento pardo o negro ; r. cefalica y bandas marginales con pelos amarillos. 14, tegumento negro con pelos negros; bandas y manchas pardo claro con pelos blancos. 15, tegumento negro; banda media y marginales con pelos amarillos; el resto con pelos negros y amarillos mezclados. 16, tegumento negro; bandas media y marginales con pelos blancos; el resto con pelos negros y blancos mezclados. 17, tegumento negro; bandas marginales con pelos amarillos; el resto con pelos negros, amarillos y rojos, mezclados. 18, tegumento negro, con pelos negros y blancos, mezclados. 19, tegumento negro, con pelos negros y amarillos, mezclados. 20, tegumento negro, con pelos negros, blancos y rojos, mezclados. GALIANO-LAS ESPECIES POLIMORFICAS DE PHIALE 71 72 THE JOURNAL OF ARACHNOLOGY SEGUNA PARTE: TAXONOMIA Metodos.— Las medidas se dan en milimetros y se han tornado segun procedimientos explicados en un trabajo anterior (Galiano 1963). Se han agregado algunas medidas de los palpos (Fig. 10) y se han hallado las relaciones entre ellas, que tienen utilidad para el reconocimiento de las especies (Tabla 1). Los colores que se describen, salvo los de P. tristis , son de ejemplares conservados en alcohol. La quetotaxia se indica segun el sistema propuesto por Platnick y Shadab (1975). Las abreviaturas son las siguientes: D = dorsal, V = ventral, p y P = prolateral, r y R = Fig. 9.- Distribution de las especies de Phiale del grupo mimica. GALIANO-LAS ESPECIES POLIMORFICAS DE PHIALE 73 retrolateral, O.L.A., O.M.A. y O.L.P. ojos laterales anteriores, medios anteriores y laterales posteriores, respectivamente. Material.— Los ejemplares de P. tristis estudiados provienen en parte de las colecciones ya existentes y en parte de las crfas realizadas en el laboratorio. Todos los especimenes de las otras especies pertenecen a las colecciones de los siguientes museos e institutos: British Museum (Natural History) (B.M.N.H.); Museum of Comparative Zoology, Cambridge, Mass., U.S.A. (M.C.Z.); Museum fur Naturkunde, Humboldt Universitat, Berlin DDR (M.N.B.); Museum National D’Histoire Naturelle, Paris, Francia (M.N.H.N.); Museu Nacional de Rio de Janeiro, Brasil (M.N.R.J.); Museu de Zoologia de la Universidade de Sao Paulo, Brasil (M.Z.S.P.); Institut Zoologique, Academie des Sciences, Warszawa, Polonia (I.Z.P.); Museo de La Plata, R. Argentina (M.L.P.); Museo Argentino de Ciencias Naturales, Buenos Aires, R. Argentina (M.A.C.N.). ESPECIES DE PHIALE DEL GRUPO MIMICA Diagnosis.— Palpo del macho robusto; femur grueso y curvo; tarso con una prolonga- cion dorsal basal conica; bulbo globoso; estilo largo, delgado, recto o levemente curvo, dirigido directamente hacia el apice del cymbium. Epigino:placa semicircular, con bolsillo mediano posterior; de cada lado, un canal curvo que termina adelante en el orificio de entrada; espermatecas esfericas, posteriores; conductos anchos de paredes gruesas, leve- mente sinuosos. Especies con polimorfismo controlado por el sexo, con machos mono- morficos y hembras polimorficas. Descripcion.— Largo total: macho 5,58-8,77; hembras 6,38-12,63. Area ocular mas ancha que larga. En los machos y en el 71,8 % de las hembras es mas ancha adelante que atras. Del resto de las hembras, 21,1 % tiene el area paralela y 7,1 % mas ancha atras. En este ultimo caso se trata siempre de ejemplares de gran tamano. Ojos de la segunda hilera mas cerca de los O.L.A. Altura del clipeo, un tercio o un cuarto del diametro de los O.M.A. ; color anaranjado, glabro en las hembras y con barba de pelos blancos o amarillos en los machos. Queliceros verticales; promargen con dos dientes, retromargen con uno: cara anterior aplanada y estriada transversalmente en los machos, con base prominente en las hembras. Largo relativo de las patas en ambos sexos IV-III-I-II. Quetotaxia mas comun: machos, femur I, II D 1-1-1, P 2, R 1-2; III D 1-1-1, P 1-2, R 1-2; IV D 1-1-1, P Fig. lO.-Medidas del palpo: a=largo del apice del cymbium, b=largo del bulbo, c=largo del estilo, d=largo del cymbium, e=largo del femur, f=ancho del femur. 74 THE JOURNAL OF ARACHNOLOGY Tabla l.-Especies de Phiale del grupo mimica. Relaciones entre las medidas del palpo: a= largo del apice del cymbium, j3=largo del bulbo, c=largo del estilo, d=largo del cymbium, e= largo del femur, f=ancho del femur, X=media, S.D.=desviacion standard, N=numero de ejemplares, L.V.=lfmites de la variation. ESPECIES RELACIONES f/e e/d c/b a/c b/d P. mimica X: 52.3 S.D. 1.32 N: 13 L.V. 50-54 X: 91.5 S.D. 4.94 N: 13 L.V. 87-105 X: 66.6 S.D. 2.67 N: 12 L.V. 62-71 X: 58.9 S.D. 3.44 N: 12 L.V. 55-65 X: 74.3 S.D. 3.61 N: 13 L.V. 71-84 P. crocea X: 48.7 S.D. 2.22 N: 9 L.V. 47-54 X: 80.6 S.D. 2.06 N: 9 L.V. 78-84 X: 87 S.D. 5.72 N: 9 L.V. 77-95 X:60.4 S.D. 2.12 N: 9 L.V. 57-63 X: 69.3 S.D. 4.87 N: 9 L.V. 64-81 P. tristis X: 47.2 S.D. 2.55 N: 21 L.V. 43-53 X: 87.8 S.D. 6.06 N: 20 L.V. 75-95 X: 78.4 S.D. 5.17 N: 22 L.V. 71-88 X: 57.7 S.D. 3.87 N: 22 L.V. 5 1-67 X: 74 S.D. 3.08 N: 22 L.V. 68-79 P. ortrudae X: 45 S.D. 3 N: 3 L.V. 42-48 X: 95.6 S.D. 0.57 N: 3 L.V. 95-96 X: 59.6 S.D. 3.21 N: 3 L.V. 56-62 X: 62.6 S.D. 4.61 N: 3 L.V. 60-68 X: 82.6 S.D. 0.57 N: 3 L.V. 82-83 P. bulbosa 55.7 - 72 63 73 1-2, R 2. Patella I P 1; II, III, IV P 1, R 1. Tibia I P 1-1, R 1-1, V 2-2-2; II P 1-1-1, R 1-1-1, V 2-2-2; III, IV D 1 , P 1-1-1, R 1-1-1, V 2-2. Metatarso I P 1, R 1, V 2-2; II P 1-1, R 1-1, V 2-2; III P 1-2, R 1-1-2, V 2-2; IV P 1-1-2, R 1-1-2, V 2-2. Hembras, femur I D 1-1-1, P 2; II D 1-1-1, P 2, R 2; IIID 1-1-1, P 1-2, R 1; IV D 1-1-1, P 1 R 1. Patella IIP 1; III, IV P 1, R 1. Tibia IP 1-1, V 2-2-2; IIP 1-1, V lr-2-2; III, IV P 1-1-1, R 1-1-1, V lp-2. Metatarso I, II V 2-2; III P 1-2, R 1-1-2, V 2-2; IV P 1-1-2, R 1-1-2, V 2-2. Pequenas variaciones aparecen en ejemplares de todas las especies. Las mas frecuentes son: machos, patella I, P 1, R 1. Tibia II P 1-1, R 1-1, V lp-2; III, IV sin dorsal;P l-l-l-l, R l-l-l-l, en dos fllas. Metatarso I P 1-1, R 1-1. Hembras, patella I P 1. Tibia IP 1; II V 2-2-2. En ambos sexos hay pequenas variaciones en las apicales de los femures. Palpo del macho con el femur relativamente corto, grueso y curvado; tibia con apofisis retrolateral conica levemente flexuosa; cymbium con una proyeccion basal conica que se apoya en el dorso de la tibia. Bulbo globoso, con una prolongation conica basal retrolateral; parte media atravesada por un pliegue oblicuo. Estilo implantado en prolateral superior, dirigido hacia el apice del cymbium. Epigino: placa limitada por un borde semicircular a menudo carenado; borde posterior levemente excavado, con bolsillo mediano. De cada lado, un canal o depresion curva termina adelante en el orificio de entrada de los conductos; a veces esta situado en el fondo de una fosa de bordes carenados. Conductos de las esper- matecas de gruesas paredes; espermatecas esfericas, ubicadas cerca del borde posterior. Patrones de diseno y colorido. Este caracter es igual en los machos de las cinco especies: prosoma negro, cubierto por pelos negros, excepto dos anchas bandas marginales GALIANO-LAS ESPECIES POLIMORFICAS DE PHIALE 75 de pelos blancos o amarillentos, que se prolongan hacia adelante y forman la barba del clipeo. Ademas hay una banda toracica media y dos manchitas entre los ojos de la segunda hilera y los O.L.P., con esos mismos pelos. Opistosoma negro, dorsalmente con pelos pardos con reflejos olivaceos. A lo largo de la lfnea media y bordeando la base y los costados, bandas de pelos blancos o amarillentos (Fig. 8 A). Vientre negro, con un triangulo basal de pelos blancos. Esternon negro, con pelos blancos en los bordes. Queliceros y laminas negros, pardos en los apices. Patas con femures negros con abun- dantes pelos blancos; patellas negras con pelos blancos basales; tibias negras con un anillo mediano pardo cubierto por pelos blancos; metatarsos y tarsos I y II, pardos, con manchitas basales de pelos blancos; metatarsos III y IV negros. Palpos con femures pardo oscuro, densamente cubiertos en el dorso y en especial en el apice, por pelos blancos; patella y tibias pardas con pelos blancos, mas abundantes en patella; cymbium negro con pelos negros. Hembras con variantes polimorficas de diseno y colorido (Figs. 8 B-P). Palpos blanquecinos con pelos blancos. Repartition.— America Central y America del Sur, hasta el norte de la R. Argentina, aproximadamente hasta los 28° S, en areas cubiertas por selva tropical y subtropical. CLAVE PARA LAS ESPECIES DE PHIALE DEL GRUPO MIMICA (EXCEPTO P. BULBOSA ) 1. Machos . 2 Hembras 5 2. Estilo curvo 3 Estilo recto . 4 3. Estilo muy largo y delgado; cymbium largo y angosto; relacion largo estilo/largo bulbo 87 ± 5.72 P. crocea Estilo mas corto y ancho; cymbium corto y robusto; relacion largo estilo/largo bulbo 78.4 ±5.17 P. tristis 4. Estilo muy corto; relacion largo estilo/largo bulbo 59.6 ± 3.21 ; relacion ancho femur/ largo femur 45 ± 3; apofisis tibial delgada, de apice agudo P. ortrudae , n. sp. Estilo relativamente mas largo; relacion largo estilo/largo bulbo 66.6 ± 2.67; relacion ancho femur/largo femur 52.3 ± 1 .32; apofisis tibial gruesa y roma P. mimica 5. Orificios de las espermatecas rodeados de gruesos rebordes quitinosos 6 Orificios sin ese reborde 7 6. Orificios muy juntos; primer tramo del conducto de la espermateca dirigido hacia adelante; borde del epigino sin carena • P. mimica Orificios muy separados; conductos dirigidos directamente hacia atras; borde del epigino carenado P. ortrudae , n. sp. 7. Orificios situados en el fondo de profundas fosas de borde carenado, en la mitad anterior del epigino P. tristis Orificios situados en un area deprimida, pero no en fosas, bien proximos al borde anterior del epigino P. crocea 76 THE JOURNAL OF ARACHNOLOGY Phiale tristis Mello-Leitao, 1945 Phiale tristis Mello-Leitao 1945:293 (hembra inmadura holotipo, de R. Argentina, provicia de Misiones, Pto. Victoria, col. Zenzes, N- 16.830; 7 machos de igual localidad y co lector deter- minados por Mello-Leitao como “tipos”, en M.L.P., examinados); Roewer 1954:1063. Phiale mutilloides Mello-Leitao 1947:27, pi. 6, fig. 13 (hembra holotipo, de Brasil, Minas Gerais: Carmo do Rio Claro, col. J. C. M. Carvalho, en M.N.R.J., examinado); Roewer 1954:1061. NUEVA SINONIMIA Phiale nigrosigillata Mello-LeitSo 1947:27, pi. 6, fig. 22 (hembra holotipo, de Brasil, Minas Gerais: Carmo do Rio Claro, col. J. C. M. Carvalho, en M.N.R.J., examinado); Roewer 1954:1062. NUEVA SINONIMIA. Phiale rubrosericea Mello-LeitSo 1947:28, pi. 6, fig. 23 (hembra holotipo, de Brasil, Minas Gerais: Carmo do Rio Claro, col. J. C. M. Carvalho, en M.N.R.J., examinado); Roewer 1954:1063. NUEVA SINONIMIA. El holotipo es un juvenil sin caracteres distintivos que permitan diferenciarlo de los inmaduros de otras especies de Phiale. Ni la descripcion ni el estudio del holotipo per- miten reconocer la especie, por lo que debria declararsela species inquirendae. No obstante, he decidido no adoptar ese criterio, por las siguientes razones: en el Museo de La Plata existen dos tubos, uno con un macho y otro con seis, ambos con etiquetas que dicen: Phiale tristis typ. “Pto. Victoria, Misiones. Zenzes col.” El primero de los ejem- plares mencionados se hallo en las colecciones del Museo de Rio de Janeiro y fue entre- gado posteriormente al Museo de La Plata. La localidad y el colector son los mismos que los del holotipo, por lo cual puede pensarse que todos estos especimenes integraban el lote original, remitido por M. Biraben a Mello-Leitao para su estudio. Es imposible saber las razones que llevaron a este autor a basar la especie en un juvenil de escaso desarrollo, cuando tenia a su disposition siete individuos adultos, pero pienso que existio cierta confusion ya que rotulo a todos como tipos y luego no los menciono en la publication. He decidido considerar que el holotipo y los ejemplares etiquetados por el autor como “tipos” pertenecen efectivamente a la misma especie. Esto permite aplicar el nombre de Phiale tristis al taxon del grupo mimica que se distribuye por el norte y noreste de la Argentina y que en este trabajo ha sido objeto de experiencias de cria. Se han observado cuatro hembras y un macho de Brasil, Goiaz asi como los tres ejemplares tipicos de P. mutilloides, nigrosigillata y rubrosericea , procedentes de Minas Gerais. El macho no se distingue de los argentinos, pero en las hembras es posible advertir una separation levemente mayor en las fosas de los epiginos (Fig. 27). Considero sin embargo, que se trata de la misma especie, por lo que se establecen las sinonimias. Descripcion.— Macho 7192 MACN. Largo total 6,10. Prosoma, largo 3,27; ancho 2,47; alto 1,47. Clipeo: alto 0,23. Area ocular: largo 1,33; ancho de la hilera anterior 1,97; de la posterior 1,87. Ojos de la segunda hilera a los O.L.A. 0.37; a los O.L.P. 0,42. Diametro de los O.M.A. 0,67. Palpo: estilo curvo, grueso (Figs. 15, 16). Color: como las especies del grupo. Hembra N? 7192 MACN. Largo total 8,78. Prosoma, largo 3,67; ancho 2,87; alto 1,73. Clipeo: alto 0,17. Area ocular: largo 1,60; ancho de la hilera anterior 2,20; de la posterior 2,13. Ojos de la segunda hilera a los O.L.A. 0,30; de los O.L.P. 0 47. Diametro O.M.A. 0,73. Epigino: placa bordeada por una carena, a menudo interrumpida en la parte anterior; en la mitad anterior del area, dos profundas fosas de borde carenado en cuyo interior se abren los orificios de entrada a los conductos (Figs. 25-27, 29). Patron de diseno y colorido: en los ejemplares criados en el laboratorio, se observaron las variantes polimorficas ilustradas en las Figs. 1 a 7. En especimenes de otras pro- cedencias, las variantes polimorficas halladas son las siguientes (Fig. 8): (1) opistosoma J; GALIANO-LAS ESPECIES POLIMORFICAS DE PHIALE 77 prosoma 5. Pata I anaranjada; tibia y apice de metatarso, negros; (2) opistosoma F pero los pelos blanquecinos en lugar de anaranjados; prosoma 14; (3) opistosoma 0, con pelos rojos en los costados, sin manchas blancas; prosoma 5; (4) opistosoma 0; prosoma 5. Pata I parda; (5) opistosoma B; (6) opistosoma G-2; prosoma 4 (? ). Pata I parda; (7) opisto- soma M-2; prosoma 15; (8) opistosoma M-3; prosoma 20; (9) opistosoma P-1, y ademas una bandita roja longitudinal; prosoma 19. Pata I parda, tibia negra; (10) opistosoma P-2; prosoma 17. Pata I parda; (11) opistosoma P-2, pero algunos pelos amarillos mezclados con los negros; prosoma 20. Pata I parda. Localidad tipica.— R. Argentina, provincia de Misiones: Puerto Victoria. Historia natural.— Ejemplares de ambos sexos se encuentran en el Parque Nacional de Iguazu, provincia de Misiones, R. Argentina, en primavera y verano. Las hembras capturadas en octubre son adultas y generalmente ya han sido fecundadas. Se hallan en la selva subtropical, en la vegetation lateral de caminos o senderos. Tejen refugios de seda blanca, mas o menos circulares, con una. dos o tres entradas. Esta estructura es igual para los ejemplares de ambos sexos de todas las edades. Dentro del refugio mudan y desovan. En cautiverio, por razones desconocidas, cambian frecuentemente de refugio, cons- truyendo uno nuevo algunos centimetros mas lejos. Luego de mudar, algunos individuos abandonan el refugio con la exuvia adentro; otros, en cambio, expulsan la exuvia y siguen viviendo en el mismo. Los refugios que se usan para desovar tienen las paredes muy gruesas, opacas, por aplicacion de mas capas de seda. Las hembras realizan un solo desove en cada refugio. Se encierran con los huevos hasta que las crias eclosionan. Cuando las aranitas abandonan el cocon, permanecen por un tiempo dentro del nido, y la madre sale para alimentarse y vuelve a entrar. El refugio es abandonado totalmente cuando todas las crias se dispersan. En el laboratorio, los prime ros ejemplares que Began a adultos son siempre machos. Sin embargo, hay algunos machos de desarrollo lento, que alcanzan la madurez despues que la mayoria de las hembras. En su ambiente natural, los juveniles pasan el invierno en pequenos refugios de seda y maduran al llegar la primavera. Distribution. -R. ARGENTINA: provincia de Misiones ; Parque Nacional Iguazu, Puerto Libertad, General Belgrano, Puerto Victoria, Puerto Esperanza, Pinalitos, Santa Maria, 2 de Mayo: Provincia de Salta ; Pocitos, Urundel; Provincia deJujuy, Ledesma. BRASIL: Goiaz; Yatai: Minas Gerais', Car mo do Rio Claro: Mato Grosso; Campo Grande, Rio Amambay. PARAGUAY: Depto. San Pedro; San Estanislao: Depto. Concepcion; Rio Aquidaban. Phiale mimica (C. L. Koch. 1846) Plexippus mimicus C. L. Koch 1846:111, pi. cccl, f. 1173 (hembra holotipo de Brasil, S. Paulo, Kat. N-° 1752 en M.N.B., examinado); 1851:52. Pardessus mimicus Peckham y Peckham 1896:36, pi. 3, f. 1-lb; 1901:302. Phiale mimica Simon 1903:695, 701, 702, 707 (n. comb.). Petrunkevitch 1911: 692. Roewer 1954: 1061. Bonnet 1958:3508. Galiano 1978: 164, f. 5, 6, 12, 13. Euophrys bella C. L. Koch 1846:203, pi. ccclxv, f. 1250 (macho holotipo de Brasil, Kat. N-° 1796 en M.N.B., examinado) NUEVA SINONIMIA Euophrys (Freya) bella C. L. Koch 1851:66 (n. subgen.). Freya bella Simon 1903:730 (n. comb.). Petrunkevitch 1911:652. Roewer 1954;1080. Bonnet 1956:1918. La redescripcion del Holotypus hembra aparecio publicada en un trabajo anterior (Galiano 1978) por lo que ahora solo se describira el macho de la especie Description.— Macho E 566 (M.Z.S.P.). Largo total 8,65. Prosoma: largo 3,67;ancho 2,93; alto 1,67. Clipeo: alto 0,27. Area ocular: largo 1,50; ancho de la hilera anterior 78 THE JOURNAL OF ARACHNOLOGY 2,10; de la posterior 2,00. Ojos de la segunda hilera, a los O.L.A. 0,40; a los O.L.P. 0,47. Diametro O.M.A. 0,73. Palpo: femur muy ancho; estilo recto y relativamente largo. (Figs. 11, 12). Patron y diseno de colorido: como en los machos del grupo. Hembras: Epigino segun figuras 23 y 31. Patron de diseno y variantes polimorficas observadas en el colorido (Fig. 8): (1) opistosoma H; prosoma 14. Pata I anaranjada; tibia, metatarso y tarso, negros; (2) opistosoma P-1; prosoma 12. Pata I pardo oscuro; metatarso amarillo; (3) opistosoma P-2; prosoma 14. Pata I negra; (4) opistosoma P-2; prosoma 16. Pata I anaranjada; femur y tarso pardo oscuro; metatarso amarillo; (5) opistosoma M-l; prosoma 18. Pata I negra; Figs. ll-16.-Palpos: 11 y 12, Phiale mimica (C. L. Koch) (holotipo de E. bella)\ 13 y 14, P. bulbosa (Cambridge) n. comb, (holotipo); 15 y 16, P. tristis Mello-Leitao. Escala=0,5mm. GALIANO-LAS ESPECIES POLIMORFICAS DE PHIALE 79 metatarso y tarso pardo claro; (6) opistosoma O pero en los costados pelos rojos; prosoma 12; (7) opistosoma O; prosoma 5. Pata I parda; femur negro; (8) opistosoma I; prosoma 15. Pata I anaranjada; tibia negra; (9) opistosoma C; prosoma 1. Pata I amarilla; femur negro; (10) opistosoma C; prosoma 6. Pata I amarilla; tibia y tarso negros; (11) opistosoma B; prosoma 1. Pata I amarilla; femur negro; (12) opistosoma F; prosoma 5. Pata I amarilla; tibia negra. Localidad tipica.— Brasil: Sao Paulo. Distribution. -BRASIL: E stado de SHo Paulo; Faz. Ubatuba, Cantareira, Cocaia: E. Guanabara ; Foresta de Tijuca: E. Bahia; Camacan, Lomanto Jr., Ju^ari;^. Espiritu Santo; Linhares, Sooretama. Phiale crocea C. L. Koch, 1846 Phiale crocea C. L. Koch 1846:194, pi. ccclxiv, f. 1242 (hembra holotipo de Brasil, Para, Kat. N-° 1623, en M.N.B., examinado); 1851:59. Simon 1903:702. Petrunkevitch 1911:689. Roewer 1954:1058. Bonnet 1958:3505. Attus obscurus Taczanowski 1872:84 (macho holotipo de Guyane Fran?aise, Cayenne, col. Jelski, N-° 61 en I.Z.P., examinado). Galiano 1966:255. ICZN 1970:100. NUEVA SINONIMIA Freya obscura Proszynski y Starega 1969:9 (n. comb.). Evophrys obscurus Taczanowski 1879:289 (1 macho N- 73, 2 machos N- 22, de Peru, Amable Marfa en I.Z.P., examinados). Petrunkevitch 1911:648. Roewer 1954:1181. Bonnet 1956:1884. Akela obscura Caporiacco 1948:723 (n. comb.). Chira luctuosa Simon 1902:52 (macho holotipo de Peru, Pebas, en M.N.H.N., examinado); 1903;747. Petrunkevitch 1911:611. Roewer 1954:1070 (Shira). Bonnet 1956:1046. Caporiacco 1954:167. NUEVA SINONIMIA Freya bulbosa Galiano 1961:186; 1963:323; 1966:255. Chickering 1946:174. Phiale mimica Petrunkevitch 1925:82. Chickering 1946:229. Phiale zonata Caporiacco 1947:31 (hembra sp. n.); 1948:708. Roewer 1954:1063. NUEVA SINONIMIA. Description.— Hembra Holotypus. Prosoma: largo 3,73; ancho 2,60. Clipeo: alto 0,17. Area ocular: largo 1,50; ancho hilera anterior 2,13; de la posterior 2,10. Ojos de la segunda hilera a los O.L.A. 0,33; a los O.L.P. 0,47. Diametro O.M.A. 0,73. Patron de diseno y colorido: opistosoma F, prosoma 5 (Fig. 8). Patas pardas, las tibias mas oscuras. Macho N-° 7194 MACN. Largo total 6,53. Prosoma: largo 3,07; ancho 2,33; alto 1,53. Clipeo: alto 0,20. Area ocular: largo 1,27; ancho hilera anterior 1,93; hilera posterior 1,80. Ojos segunda hilera a los O.L.A. 0,33; a los O.L.P. 0,37. Diametro O.M.A. 0,63. Palpo: estilo curvo, muy largo y delgado (Fig. 17, 18). Color: como las especies del grupo. Hembras. Epigino segun figuras 22 y 28. Patron de diseno y colorido, variantes polimorficas observadas (Fig. 8): (1) opistosoma C; prosoma 1. Pata I anaranjada; femur negro; tarso pardo; (2) opistosoma G-2; prosoma 4. Pata I anaranjada; tibia parda; (3) opistosoma G-2; prosoma 9. Pata I pardo oscuro; metatarso pardo claro; (4) opistosoma G-3; prosoma 8. Pata I negra; metatarso amarillo; (5) opistosoma E; prosoma 9. Pata I pardo claro; femur y tarso, negros; (6) opistosoma J; prosoma 5. Pata I anaranjada; tibia y tarso, negros; (7) opistosoma B; prosoma 9; (8) opistosoma B; prosoma 2. Pata I amarilla; femur y tarso, negros; (9) opistosoma B; prosoma 12. Pata I pardo claro; femur y tarso, negros; (10) opistosoma B; pero con las bandas muy angostas; prosoma 9. Pata I anaran- jada; femures y tarsos, pardos; (11) opistosoma B; prosoma 4; (12) opistosoma B; pero con pelos anaranjados; prosoma 4. Pata I amarilla; tibia negra; (13) opistosoma N, con las manchas laterales pequenas; prosoma 14. Pata I pardo oscuro; (14) opistosoma K; pro- soma 12. Pata I pardo oscuro; metatarso y tarso pardo claro; (15) opistosoma K, pero sin las bandas basales laterales; prosoma 14. Pata I pardo oscuro; metatarso pardo claro; (16) 80 THE JOURNAL OF ARACHNOLOGY opistosoma P-3; prosoma 5; (17) opistosoma F; prosoma 4. Pata I anaranjada; tibia negra; (18) opistosoma M-l, con dos pares de manchas apicales de pelos anaranjados; prosoma 12. Pata I anaranjada; femur y tarso, negros; (19) opistosoma M-2; prosoma 19. Pata I pardo oscuro; metatarso amarillo; (20) opistosoma D; prosoma 8. Pata I negra; metatarso y tarso pardo claro; (21) opistosoma 0, pero los costados con pelos blancos y negros; prosoma 12. Pata I pardo oscuro; (22) opistosoma L; prosoma 12. Pata I parda; tibia negra. Localidad tipica.— Brasil, Para. Observation.— En publicationes anteriores (Galiano 1961, 1963, 1966) antes de haber examinado el holotipo de Cyrene bulbosa Cambridge, 1901, crei que se trataba de un Figs. 17-21. -Palpos: 17 y 18, Phiale crocea C. L. Koch; 19, 20 y 21, P. ortrudae n. sp. (holotipo). Escala=0,5mm. GALIANO-LAS ESPECIES POLIMORFICAS DE PHIALE 81 sinonimo de Attus ob scurus. Segun puede verse en las Figs. 13, 14, 17 y 18, se trata de especies diferentes. La dispersion de los machos identificados como Attus obscurus se superpone con la de las hembras de Phiale crocea por lo que no vacilo en considerarlos sinonimos. Chira luctuosa Simon, debe tambien ubicarse en la sinonimia de esta especie. En 1946, Chickering creyo reconocer en los machos de Panama a Freya bulbosa (Cam- bridge) y en las hembras a Phiale mimica (Koch). Ob servo que entre su material habia 23 Figs. 22-27 .-Epiginos: 22, P. crocea ; 23, P. mimica ; 24, P. ortrudae n. sp. (paratipo); 25, P. tristis', 26, P. tristis, variation; 27, P. tristis (holotipo de P. mutilloides). Escala=0,5mm. 82 THE JOURNAL OF ARACHNOLOGY ejemplares con un patron de diseno y colorido tipico de mimica y d egratiosa ademas de formas intermedias, sin que existieran diferencias significativas en las estructuras funda- mentals, por lo que concluyo que eran sinonimos. Sospecho tambien que P. crocea pertenecia a este taxon. Segun demuestra el estudio de los ejemplares tipicos de todas las especies mencionadas Attus obscurus y Freya bulbosa son distintas y los ejemplares de Panama estudiados por Chickering son Phiale crocea. Los especimenes de Bolivia se asignan a P. crocea con ciertas dudas, pues los epiginos presentan caracteres intermedios con P. tristis. Distribution. -BRASIL: Territorio de Amapa; Serra do Navio: Para ; Belem, Santarem. GUYANA: Kartabo. GUYANE FRANCAISE: Cayenne. PERU: San Ramon, Tingo Maria, Amable Maria. Iquitos. VENEZUELA: Territorio Federal de Amazonas’, Atures: Carabobo’, San Esteban. ECUADOR: Napo. COLOMBIA: Boyaca. TRINIDAD: Port of Spain, Simla. PANAMA: Canal Zone. BOLIVIA: Dto. Cochabamba : R. Chapare. Phiale bulbosa (F. 0. P. Cambridge, 1901), n. comb. Cyrene bulbosa F. O. P. Cambridge 1901:231, pi. 18, f, 16, 16a-d (macho holotipo de Panama, Bugaba, col. Champion, N- 372, en B.M.N.H., examinado). Freya bulbosa Petrunkevitch 1911:653; 1925:81. Roewer 1954: 1080. Bonnet 1956:1919. Figs. 28-31. -Espermatecas y conductos: 28, P. crocea’, 29, P. tristis’, 30, P. ortrudae n. sp. (paratipo); 31, P. mimica. Escala=0,25mm. GALIANO-LAS ESPECIES POLIMORFICAS DE PHIALE 83 Description.— Macho Holotypus. Prosoma: largo 2,83; ancho 2,17; alto l,37.Clipeo: alto 0,17. Area ocular: largo 1,30; ancho hilera anterior l,83;hilera posterior 1,68. Ojos de la segunda hilera a los O.L.A. 0,33; a los O.L.P. 0,40. Diametro O.M.A. 0,67. Palpo: estilo recto, levemente mas angosto en su parte media (Figs. 13,14). Patron de diseno y colorido: como en las especies del grupo. Localidad tipica.— Panama: Bugaba. Observation. -El tipo es el unico ejemplar hasta ahora conocido. La estructura del palpo indica una gran proximidad con P. mimica, de la cual se diferencia por la forma particular del estilo. Existen tres posibilidades: (a) que se trate de una especie valida, de la cual no se ha colectado hasta ahora mas que un ejemplar, simpatrida con P. crocea en el borde de su area de distribution; (b) que sea un individuo d eP. mimica algo anormal, en cuyo caso el area de distribution se veria grandemente extendida, con lagunas intermedias y (c) que sea un ejemplar algo anormal de P. mimica y que hay a un error en la indication de la procedencia. Estas dudas podran clarificarse cuando se obtenga mas material de e studio. Phiale ortrudae, nueva especie Diagnosis.— se diferencia de las otras especies del grupo por tener el estilo mas corto en relation a los largos del bulbo y cymbium; las hembras se distinguen por la gran distancia entre los orificios de entrada a los conductos de las espermatecas. Description.— Macho Holotypus. Largo total 7,45. Prosoma: largo 3,20; ancho 2,50; alto 1,60. Clipeo: alto 0,23. Area ocular: largo 1,35; ancho hilera anterior 1,90; hilera posterior 1,77. Ojos de la segunda hilera a los O.L.A. 0,37; a los O.L.P. 0,43. Diametro O.M.A. 0,63. Palpo: femur relativamente delgado; bulbo muy desarrollado; estilo recto, corto; apoflsis tibial delgada, de apice agudo (Figs. 19-21). Patron de diseno y colorido: como las especies del grupo. Hembra Paratypus: Largo total 9,71. Prosoma: largo 3,53; ancho 2,73; alto 1,67. Clipeo: alto 0,17. Area ocular: largo 1,47; ancho hilera anterior 2,03; hilera posterior 2,00. Ojos de la segunda hilera a los O.L.A. 0,37; a los O.L.P. 0,47. Diametro de O.M.A. 0,67. Epigino: orificios de las espermatecas rodeados de gruesos rebordes quitinosos; muy separados entre si; conductos dirigidos directamente hacia atras (Figs. 24, 30). Patron de diseno y colorido, variantes polimorficas observadas (Fig. 8): (1) opistosoma M-l; pro- soma 18. Pata I pardo oscuro, metatarso amarillo; (2) opistosoma M-l ; prosoma 16. Pata I anaranjada; femur y tarso negros; metatarso amarillo; (3) opistosoma M-2; prosoma 19. Pata I pardo oscuro, metatarso amarillo; (4) opistosoma F; prosoma 5. Pata I anaranjada; tibia y tarso negros. Localidad tipica.— Ecuador: Provincia de los Rios: Quevedo. Material estudiado: un macho Holotypus N- 7189 MACN; dos hembras Paratypi Np 7190 MACN, col. Fritz IV-1976. Una hembra Paratypus Np 7191 MACN, de Ecuador: Prov. de los Rios; Pichilingue, col. A. Martinez V-1976. Dos machos, tres hembras y un pullus Paratypi (M.C.Z.) de Ecuador; Pichincha, S. Domingo, col. S. & J. 18-30-V-1975. AGRADECIMIENTOS Por el prestamo del material tipico y de colecciones indeterminadas, expreso mi re- conocimiento a las siguientes personas: Dr. M. Vachon, Dr. H. W. Levi, Sr. F. R. Wanless, Dr. M. Moritz, Dra. Ana Timotheo da Costa, Dr. P. Vanzolini, Dra. L. Neme, Dr. W. 84 THE JOURNAL OF ARACHNOLOGY Starega, Dr. J. Proszynski, Dra. Olga Blanco. Tambien quedo reconocida a los entomo- logos G. Williner, A. Martinez y M. Fritz, por la cesion de ejemplares por ellos colectados; a la Srta. Elvira Buono por el acabado en tinta de algunos dibujos. LITERATURA CITADA Blandin, P. 1976. Etudes sur les Pisauridae africaines, VI. Definition des genres Pisaura Simon, 1885, Pisaurellus Roewer, 1961, Afropisaura n. gen.et mise au point sur les especes des genres Afropi- saura et Pisaurellus (Araneae, Pisauridae, Pisaurinae). Rev. Zool. afr., 90(4):9 17-939. Blandin, P. 1977. Etudes sur les Pisauridae africaines. VIII. Les genres Chiasmopes Pavessi, 1883 et Rothus Simon, 1898 (Araneae, Pisauridae, Pisaurinae). Rev. Zool. afr., 9 1 (3) :5 38-557 . Bonnet, P. 1955-1958. Bibliographia Araneorum, 2(2-4):919-4230. Cambridge, F. O. P. 1898-1903. Biologia Centrali-Americana, Arachnida, 2:1-610. Caporiacco, L. 1947. Diagnosi preliminari di specie nuove di Archnidi della Guiana Britannica raccolte dai Professori Beccari e Romiti. Monit. zool. ital., 56(l-6):20-34. Caporiacco, L. 1948. Arachnida of British Guiana collected in 1931 and 1936 by Professors Beccari and Romiti. Proc. Zool. Soc. London, 118:607-747. Caporiacco, L. 1954. Araignees de la Guyane Fran§aise du Museum d’Histoire Naturelle de Paris. Comment., 16(3):45-193. Chickering, A. 1946. The Salticidae (Spiders) of Panama. Bull. Mus. comp. Zool. Harvard, 97:1-474. Emerit, M. 1969. Contribution a l’etude des Gasteracanthes (Araneides, Argiopides) de Madagascar et des lies voisines. These Univ. Montpellier AO-2888, p.1-434. Emerit, M. 1973. Contribution a la connaissance des Araneidae Gasteracanthinae du sud-est africain: les Gasteracanthes du Natal Museum. Ann. Natal Mus., 21(3):675-695. Ford, E. B. 1940. Polymorphism and Taxonomy, pp. 493-513. In The New Systematics (J. Huxley, ed.) Oxford Univ. Press. London. Ford, E. B. 1945. Polymorphism. Biol. Rev. Cambridge Philos. Soc., 20:73-87. Ford, E. B. 1953. The Genetics of Polymorphism on the Lepidoptera. Adv. Genetics, 5 :43-87. Galiano, M. E. 1961. Revision de genero Chira Peckham, 1896 (Araneae, Salticidae). Comun. Mus. Argent. Cienc. Nat. Zool., 3(6): 159-188. Galiano, M. E. 1963. Las especies americanas de aranas de la familia Salticidae descriptas por Eugene Simon. Redescripciones basadas en los ejemplares tipicos. Physis, 23(66):273-470. Galiano, M. E. 1966. Attus obscurus Taczanowski, 1872 (Araneae); proposed suppression under the Plenary Powers in favour of Cyrene bulbosa Cambridge, 1901. Bull. zool. Nomencl., 23(5):255. Galiano, M. E. 1978. Revision del genero Phiale Koch, C. L., 1846 (Araneae, Salticidae). I. Redescrip- cion de Phiale gratiosa, P. mimica y P. rufoguttata. Physis Sec. C, 37(93):161-167. Hill, D. E. 1978. Some unusual Phidippus audax from Northern Florida. Peckhamia, 1 (4) : 7 1 -73 . International Commission on Zoological Nomenclature (ICZN). 1970. Attus obscurus Taczanowski, 1872 (Araneae): refusal to suppress under the Plenary Powers. Bull. zool. Nomencl., 27(2):100. Kolosvary, G. 1932. Beitragezu den Dessinvariationen der Spinnen. Zool. Anz., 100:192-198. Koch, C. L. 1846. Die Arachniden, 13:1-234. Koch, C. L. 1851. Uebersicht des Arachnidensy stems, 5:1-104. Levi, H. W. 1957. The spider genera Enoplognatha, Theridion, and Paidisca in America north of Mexico (Araneae, Theridiidae). Bull. Am. Mus. Nat. Hist., 1 12(1): 1-123. Levi, H. W. 1959. Problems in the spider genus Steatoda (Theridiidae). System. Zool., 8(3) : 107-1 16. Levi, H. W. 1975a. The American Orb-weaver genera Larinia, Cercidia and Mangora north of Mexico (Araneae, Araneidae). Bull. Mus. comp. Zool. Harvard, 147(3): 100-135. Levi, H. W. 1975b. Additional notes on the Orb-weaver genera Araneus, Hyposinga and Singa north of Mexico (Araneae, Araneidae). Psyche, 82(2): 265-274. Levi, H. W. 1977. The American Orb-weaver Cyclosa, Metazygia and Eustala north of Mexico (Araneae, Araneidae). Bull. Mus. comp. Zool. Harvard, 148(3):61-127. Mello-Leitao, C. 1945. Aranas de Misiones Corrientes y Entre Rios Rev. Mus La Plata (N.S.) Zool., 4(29):213-302. Mello-LeitSo, C. 1947. Aranhas do Carmo do Rio Claro (Minas Gerais) coligidas pelo naturalista Jose C.M. Carvalho. Bol. Mus. Nac. Rio de Janeiro, 80:1-34. Peckham, G. W. and E. G. Peckham. 1896. Spiders of the Family Attidae from Central America and Mexico. Occas. Pap. Nat. Hist. Soc. Wisconsin, 3:1-101. GALIANO-LAS ESPECIES POLIMORFICAS DE PHIALE 85 Peckham, G. W. and E. G. Peckham. 1901. Spiders of the Phidippus group of the family Attidae. Trans. Wisconsin Acad. Sci. Arts Lett., 13(l):282-358. Petrunkevitch, A. 1911. A Synonymic Index-Catalogue of Spiders of North, Central and South Amer- ica with all adjacent Islands, Groenland, Bermuda, West Indies, Terra del Fuego, Galapagos, etc. Bull. Am. Mus. Nat. Hist., 29:1-270. Petrunkevitch, A. 1925. Arachnida from Panama. Trans. Connecticut Acad. Arts Sci., 27:51-248. Platnick, N. I. and M. V. Shadab 1975. A revision of the spider genus Gnaphosa (Araneae, Gnaphosidae) in America. Bull. Am. Mus. Nat. Hist., 155(1 ) : 1-66. Proszynski, J. and W. Starega 1969. Comment on the proposed suppression of Attus obscurus Taczanowski, 1872. Bull. zool. Nomencl., 26(1):9. Roewer, C. F. 1954. Katalog der Araneae, 2b: 927-1 751. Seligy, V. L. 1969. Biological aspects of pigment variation in the spider Enoplognatha ovata (Clerck) (Araneae, Theridiidae). Canadian J. Zool., 46(6): 1103-1 105. Seligy, V. L. 1971. Postembryonic development of the spider Enoplognatha ovata (Clerck) (Araneae, Theridiidae). Zool. J. Linn. Soc., 50(1):21-31. Simon, E. 1897-1903. Histoire Naturelle des Araignees, 2:1-1080. Simon, E. 1902. Descriptions d’Arachnides nouveaux de la Famille des Salticidae (Attidae). Ann. Soc. entomol. Belgique, 46:24-54. Taczanowski, L. 1872. Les araneides de la Guyane Fransaise. Horae Soc. entomol. Ross., 8:32-132. Taczanowski, L. 1879. Les araneides du Perou. Bull. Soc. Imp. Nat. Moscou, 53(4):278-374. Taylor, B. B. and W. B. Peck. 1975. A comparision of northern and southern forms of Phidippus audax (Hentz)(Araneae, Salticidae). J. Arachnol., 2:89-99. Tweedie, M. W. F. 1970. A case of balanced polymorphism in a spider. Entomol. Gaz., 21 :50. Manuscript received March 1980 Jackson, R. R. 1981. Nest-mediated sexual discrimination by a jumping spider ( Phidippus fohnsoni). J. Arachnol., 9:87-92. NEST-MEDIATED SEXUAL DISCRIMINATION BY A JUMPING SPIDER (PHIDIPPUS JOHNSON!) Robert R. Jackson1 Department of Zoology, University of California Berkeley, California 94720, U.S.A. ABSTRACT Empty nests of conspecific females elicit courtship behavior from males of Phidippus fohnsoni. Males discriminate between nests of adult males and adult females and also between nests of subadult males and subadult females. INTRODUCTION Spiders are the major example in the animal kingdom of adaptive radiation in the use of silk, and it is not surprising that communication is found among its functions. The salticid spiders, however, are unlike most spiders in having highly developed vision (Land 1972) on which they rely in most aspects of their behavior, and the communicatory behavior most commonly associated with spiders of this family consists of dancing and specialized movements and postures of the legs, pedipalpi, and abdomen (Crane 1949). Communication involving the use of silk in this family has received little attention. However, Bristowe (1958) noted that the males of various salticid species show excited “signs of awareness” when they touch unoccupied nests built by conspecific females. Males of Phidippus fohnsoni Peckham and Peckham, a common species in western North America, employ visually mediated courtship when they encounter adult females in the presence of adequate light (Jackson 1977 a,b, 1978 a); however, many salticids, including P. fohnsoni, build silken nests under rocks, in hollow reeds, and in other dimly lit locations and use these for molting, ovipositing, and passing periods of inactivity (Jackson 1979). If a male P. fohnsoni encounters an adult female inside her nest, he employs vibratory courtship which does not depend on vision. If the female inside the nest is a subadult, the male first courts, then spins a second chamber on the female’s nest and cohabits with her until she matures (Jackson 1977). Males of various non-salticid spiders have been shown to begin courtship in response to contact with draglines, webs, or nests and to make species- and sex-discriminations on the basis of the silk after contact (e.g. Tietjen 1977, Jackson 1978b). This paper is concerned ‘Present address: Department of Zoology, University of Canterbury, Christchurch 1, New Zealand. 88 THE JOURNAL OF ARACHNOLOGY with similar adaptations related to communication and the use of silk in P. johnsoni. In an earlier study (Jackson 1976a) it was shown that male P. johnsoni respond to empty nests of adult females with elements of behavior that normally occur during courtship and that they discriminate between the nests of adult conspecific females and the similarly sized nests of Herpyllus hesperolus (Chamberlin), a sympatric gnaphosid spider that is a potential predator of P. johnsoni. Two questions will be considered in this paper. Do adult males discriminate between the nests of adult females and adult males? Do adult males discriminate between nests of subadult females and subadult males? METHODS Spiders.— Spiders were collected as immatures from Til den Regional Park, Oakland, California. Each adult male and female was 5 to 30 days post-maturity at the time of the tests. All females were virgins. None of the spiders had been involved in interactions with other spiders in the laboratory previous to the tests. Maintenance and Apparatus.— Details of maintenance are provided elsewhere (Jackson 1974, 1976b). Each spider was kept individually in a transparent plastic cage (10 x 10 x 6.5 cm) with a ventilation hole covered by a metal screen and two 4.5-cm-diameter holes plugged with corks (see Jackson 1978a, Fig. 1). Each spider spun at least one nest fastened to one of the two corks, and the opposite cork-hole provided access to the interior of the cage without damaging the nest. Testing Procedure.— All tests took place within the first 4 hr. after the lights came on in the laboratory (12L: 12D, lights on at 0900 hr; temp., 24° C). Each spider was forced from its nest with a camel hair brush and removed from the cage 2 to 8 min before introduction of the test male. Any additional nests and any living flies in the cages were removed. All nests were relatively dense (see Jackson 1979) and were 7 to 30 days old. Time elapsing between introduction of the male and his first touching the nest was recorded. A description of his behavior for 30 min following contact was recorded on a tape recorder with a metronome providing a time base (one beat per sec). The nest used in each test came from a different spider. Each male was tested with one nest on one day and on the following day with a different one. No male was tested with more than one pair of nests. One nest was from a female and the other from a male, each either adult or subadult. One half of the males were tested on the first day with nests of females; the other half, first with nests of males. Twelve males were tested with nests of adults and 18 with nests of subadults. A random numbers table (Rohlf and Sokal 1969) was used to assign males to groups and particular nests to particular males. No spider provided nests for more than one test. Some of the males were used both for testing and for nest production, but they were never tested with their own nests. Occurrences of behavior in the pairs of tests were compared using McNemar tests with Yates correction; latencies and durations were compared using two-tailed Wilcoxon signed-ranks tests (Sokal and Rohlf 1969). Latencies were compared only for those males that performed the behavior in question during both tests. Data in text are means ± S.D’s when normally distributed; otherwise, medians followed by maxima are provided. Elements of Behavior.— The behavior of males in the presence of occupied nests has been described elsewhere (Jackson 1977a), and similar behavior occurred at empty nests. Brief descriptions of the important behaviors are provided here. Twitch abdomen. Rapid, low amplitude, up and down movements of the abdomen. JACKSON-NEST DISCRIMINATION BY SALTICID 89 Table 1. -Number of males of Phidippus johnsoni that performed different elements of behavior after contact with empty nests. A: nests of adults. S: nests of subadults. Behavior Female only Nests at which Behavior was Performed Male only Both Neither Abdomen Twitch A 7 0 3 2 S 14 0 1 3 Vibrate A 5 0 2 5 S 5 1 0 12 Tug A 2 5 3 2 S 9 3 2 4 Probe A 0 0 12 0 S 3 0 15 0 Enter Nest A 3 0 9 0 S 1 1 16 0 Probe. Pushing and pulling on the nest with the first pair of legs. Tug. Up and down movements of the cephalothorax while the chelicerae grip the silk. Vibrate. Very rapid, low amplitude, up and down movements with the forelegs on the silk. In other studies (Jackson 1977a) twitching of abdomen, probing and vibrating were characteristic of interactions between conspecific spiders. Tugging occurred sometimes while spiders were alone and spinning in their own nests, but it was more prevalent in interactions with other spiders. RESULTS All males contacted nests 1-28 min after introduction into the cages, and there was no difference in the latency to contact between nests of females and those of males (latency for each male to contact nest of female minus that for the same male to contact nest of male: 1.6 ± 9.16 min for nests of adults; 0.9 ± 10.32 min for nests of subadults), providing no evidence of attraction to nests by airborne pheromones. Each element of behavior occurred sometimes at nests of each sex-age class (Table 1). However, males more often performed abdomen twitching at nests of females than at those of males (nests of adults: x2 = 5.143, P < 0.025; subadults: x2 = 12.071, P < 0.005). Abdomen twitching, vibrating, tugging and probing tended to occur either soon after the male touched the nest or not at all (latencies in sec from nest-contact until behavior was performed: abdomen twitch, 18, 54; vibrate, 33, 217; tug, 28, 584; probe, 2, 83). The durations (in sec) of these behaviors tended to be brief (abdomen twitch, 14, 144yibrate, 4, 37;tug, 10, 110;probe, 9, 67; cases in which the behavior failed to occur deleted). Times not engaged in these activities were occupied with grooming, walking, spinning, pivoting, or simply standing inactive. 90 THE JOURNAL OF ARACHNOLOGY Spiders that entered nests did so soon (35 sec, 395 sec) after contacting the nest. Sixteen males each entered the nests of both (a) a subadult female and (b) a subadult male, and latencies were greater when the nests were ones of (a) rather than (b) (latency in sec of (a) - (b): 43, 388; Wilcoxon test, T§ = 6, P < 0.01). Males also probed longer at nests of females than at nests of males (durations in sec of probing at nests of females minus those at nests of males: 10, 41 for nests of adults, T§ = 9; P < 0.02; 10, 64 for nests of subadults, T§ = 1 1 .5, P < 0.01). Twelve males each first touched one of the two nests with which they were tested in the vicinity of the door (an elastic slit-like opening to the nest) and the other nest elsewhere; and latencies to probe for these males were less when first contact was near the door (latency to probe in sec after contact away from door minus latency after contact near door: 3, 28; T§ = 0, P < 0.01). Also, whenever probing occurred, it was concentrated in the vicinity of the door (duration of probing near door/total duration of probing x 100%: 64% ±41.3%). DISCUSSION The behavior of male P. johnsoni is adapted to mating with females that they locate in nests. One line of evidence for this is the different motor patterns and sensory modalities involved depending on whether the female is inside or outside her nest (Jackson 1977a, b). The present study provides further evidence. The nest alone elicits courtship from the male and provides the male with information concerning the sex of the previous occupant. Although empty nests elicit brief courtship behavior, probably the presence of the nest occupant is necessary in order to sustain the male’s behavior for more than a few minutes, since courtship tended to last many minutes (mean, 16 min; max., 3 hr 44 min; Jackson 1978a) when females were inside nests. Males seem to be adapted to briefly “announcing” themselves (courting) when they contact nests of conspecific spiders and to desisting when they do not receive an “answer” (response of the female). It seems likely that chemical stimuli were involved in the discriminatory behavior of male P. johnsoni since in other groups of spiders there is evidence of pheromones associated with silk (Kaston 1936, Millot 1946, Tietjen 1978). There were no obvious differences in the shape, size, or other gross structural characteristics of the nests used in this study, suggesting that tactile discriminations of gross structural differences were not involved. Visual discrimination of either gross or fine differences also seems unlikely as an important factor since nests in nature were located in places with little ambient light; but it is possible that males discriminated fine tactile differences upon contact with nests. In this study and in ones in which the nests were occupied (Jackson 1977a), probing was concentrated in the vicinity of the door; and probing occurred more quickly when males contacted nests near the doors. Perhaps tactile stimuli associated with the opening in the nest facilitate probing, although concentration of pheromones around the door is another possibility. The communicatory behavior of P. johnsoni is more complex than that normally associated with spiders. One aspect of this complexity is the use of alternative tactics for females inside vs outside nests and for adults vs subadults. Another factor is that the rules describing when different behaviors occur do not seem simple. For example, some behaviors performed by males are also performed by females and immatures, and some of JACKSON-NEXT DISCRIMINATION BY SALTICID 91 the male’s behaviors are the same when interacting with males or females encountered in nests. Since the manner in which behaviors are segregated according to sex-age class is quantitative rather than qualitative (Jackson 1977a), the task of determining the functions and the message-meaning relationships of the displays exchanged between spiders and the signals associated with nests will be difficult. The fact that the nests of the sympatric gnaphosid spider H. hesperolus failed to elicit vibratory courtship behavior from males of P. johnsoni (Jackson 1976a) indicates that reliable species-identifying information is provided through the nests, at least with respect to these two species. The occurrence of many of the same elements of behavior at nests of both males and females of conspecifics, however, might be related to the presence of less reliable sex-identifying information or to behaviors having adaptive significance during both male-male and male-female interactions. Entry into the nest by a male is likely to elicit departure by a subadult female and decrease the male’s chances of cohabiting (Jackson 1977a). The adaptive significance of the longer latencies before entering nests of subadult females compared to subadult males may be related to this. Crane (1949) found evidence suggesting that airborne pheromones emanating from females of Corythalia, Phiale, and other neotropical salticids lowered the males’ thres- holds for performance of displays. This raises the possibility that residual airborne pheromones in the cage after removal of the females might have affected the manner in which males of P, johnsoni responded to nests since the maximum time that elapsed between removal of spiders from their nests and when the test males touched the nests was 22 min in this study. However, in the earlier study (Jackson 1976a), a full day elapsed between removal of the females and testing. Also, the cages had ventilation openings. Consequently, chemotactic or tactile stimuli would seem more important than residual airborne pheromones in eliciting responses from males in this study. ACKNOWLEDGEMENTS For valuable discussions during the early phases of this study, I thank Roy Caldwell. I thank Lawrence Field, Mary Catharine Vick, Jerome Rovner and William Eberhard for comments on the manuscript. REFERENCES Bristowe, W. S. 1958. The world of spiders. London: Collins. Crane, J. 1949. Comparative biology of salticid spiders at Rancho Grande, Venezuela. Part IV. An analysis of display. Zoologica, 34:159-214. Jackson, R. R. 1974. Rearing methods for spiders. J. Arachnol., 2:5 3-56. Jackson, R. R. 1976a. Predation as a selection factor in the mating strategy of the jumping spider Phidippus johnsoni (Salticidae, Araneae). Psyche, 83:243-255. Jackson, R. R. 1976b. The evolution of courtship and mating tactics in a jumping spider, Phidippus johnsoni (Araneae, Salticidae). Ph.D. dissertation. University of California, Berkeley. Jackson, R. R. 1977a. An analysis of alternative mating tactics of the jumping spider Phidippus johnsoni (Araneae, Salticidae). J. Arachnol., 5:185-230. Jackson, R. R. 1977b. Courtship versatility in the jumping spider, Phidippus johnsoni (Araneae: Salticidae). Anim. Behav., 25:953-957. Jackson, R. R. 1978a. The mating strategy of Phidippus johnsoni (Araneae, Salticidae): I. Pursuit time and persistence. Behav.Ecol. Sociobiol., 4:123-132. 92 THE JOURNAL OF ARACHNOLOGY Jackson, R. R. 1978b. Male mating strategies of dictynid spiders with differing types of social organi- zation. Symp. Zool. Soc. London, 42:79-88. Jackson, R. R. 1979. Nests of Phidippus johnsoni (Araneae, Salticidae): characteristics, pattern of occupation, and function. J. Arachnol., 7:47-58. Kaston, B. J. 1936. The senses involved in the courtship of some vagabond spiders. Entomol. Amer., 16:97-167. Land, M. F. 1972. Mechanisms of orientation and pattern recognition by jumping spiders (Salticidae), pp. 231-247. In Information processing in the visual systems of arthropods (ed. by R. Wehner). Berlin: Springer-Verlag. Millot, J. 1946. Sens chemique et sens visuel chez les araignees. Ann. Biol. Anim., 22:1-21. Rohlf, F. J. and R. R. Sokal, 1969. Statistical Tables. San Francisco: Freeman. Sokal, R. R. and F. J. Rohlf, 1969. Biometry. San Francisco: Freeman. Tietjen, W. J. 1977. Dragline-following by male lycosid spiders. Psyche, 84:165-178. Tietjen, W. J. 1978. Is the pheromone of Lycosa rabida (Araneae, Lycosidae) deposited on a substratum? J. Arachnol., 6:207-212. Manuscript received November 1979, revised March 1980. Sissom, W. D. and O. F. Francke 1981. Scorpions of the genus Paruroctonus from New Mexico and Texas (Scorpiones, Vaejovidae). J. Arachnol., 9:93-108. SCORPIONS OF THE GENUS PARUROCTONUS FROM NEW MEXICO AND TEXAS (SCORPIONES, VAEJOVIDAE) W. David Sissom1 and Oscar F. Francke Departments of Biological Sciences and Entomology Texas Tech University, Lubbock, Texas 79409 ABSTRACT The distribution of Paruroctonus gracilior (Hoffmann) and P. utahensis (Williams) in New Mexico and Texas is confirmed and new state records of both are established. P. aquilonalis (Stahnke) is established as a junior synonym of P. boreus (Girard), and state records of both the junior and senior synonyms are considered to be based on misidentifications. Two new species of the genus are de- scribed: P. williamsi from the Big Bend region in Texas andP. pecos from sand dunes in southeastern New Mexico. INTRODUCTION The genus Paruroctonus Werner, 1934, occurs from southern Canada, across the west- ern United States and south into Mexico. Throughout this range 23 nominal species have been recognized, and four species have been reported from New Mexico and Texas. There has been considerable confusion, however, concerning the taxonomic status of popula- tions from New Mexico and Texas. Some of the confusion is the result of records published before the genus was widely recognized as a valid taxon, additional confusion is due to erroneous identifications, and finally the inadequate description of at least one species has also contributed significantly. This contribution, based on the study of thousands of specimens representing all nominal taxa of the genus Paruroctonus described to date, aims to clarify the taxonomic status of populations from New Mexico and Texas. One nominal species is relegated to synonymy and state records of both the senior and junior synonym are established as misidentifications. Distribution records of two species are confirmed and expanded, and two new species are described. Paruroctonus boreus (Girard, 1854) Type data.— Type specimen of uncertain sex from the valley of the Great Salt Lake, Utah, collected by Howard Stansbury. Presumed still to be deposited in the United States National Museum (Gertsch and Soleglad 1966); not examined. Present address: Department of General Biology, Vanderbilt University, Station B, Box 1812, Nash- ville, Tenn. 37235. 94 THE JOURNAL OF ARACHNOLOGY Remarks.— This species has the northernmost distribution of the genus. Centered in the Great Basin Desert of North America, it extends northward into British Columbia, Al- berta, and possibly Saskatchewan in Canada; eastward to North Dakota, South Dakota, and Nebraska; westward to Washington, Oregon, and California, and southward into northern Arizona and southwestern Colorado in the United States. One immature individual has been reported from Eagle Pass, Maverick Co., Texas, by Girard (1854). We have not seen any P. boreus from Texas, and consider this record a misidentification. (It could be Vaejovis coahuilae Williams, which is commonly found in the area and has color markings reminescent of immature P. boreus .) Paruroctonus boreus was reported from White Sands National Monument, Otero and Dona Ana counties, New Mexico, by Bugbee (1942). We have not seen any specimens referable to P. boreus from New Mexico; samples from White Sands National Monument belong to Paruroctonus utahensis (Williams), which is discussed below. Therefore, we consider Bugbee’s report of P. boreus from New Mexico to be based on a misidentifi- cation. The presence of P. boreus in northwestern New Mexico is very likely however, espe- cially in San Juan and McKinley counties. It has been found in Apache Co., Arizona, Montezuma Co., Colorado, and San Juan Co., Utah, which are three of the counties (and states) adjoining each other at Four Corners. Paruroctonus aquilonalis (Stahnke, 1940) Type data.— Holotype male from 30 miles south of the Grand Canyon, Arizona, 8 August 1938 (Kay Anderson). Deposited in the collection of H. L. Stahnke; examined. Remarks.— This nominal taxon, originally described from northern Arizona, has caused considerable taxonomic confusion over the years. The original description (Stahnke 1940: 101), a short excerpt from a dissertation (Stahnke 1939), reads as follows: “Vejovis aquilonalis. First segment of cauda has no distinct, granular inferior keels. Carapace shorter than fifth caudal segment and slightly shorter than movable finger of pedipalp. Dorsum of a uniform orange-brown color. The specimen, a male, was taken 37 miles south of the Grand Canyon on highway 64.” The discrepancy in the type locality data seems to be a typographical error, for the label with the holotype clearly reads “30 mi. South.” According to Stahnke (1939), P. aqui- lonalis differs from P. boreus in two characters: the relative proportions of the carapace and movable finger of the pedipalps and in color. While Stahnke (1939, 1940) clearly indicates that in P. aquilonalis the carapace is slightly shorter than the movable finger of the pedipalps, the measurements he reported contradict this. In Stahnke’s dissertation (1939:75), the measurements given are: cara- pace length 4.4 mm , movable finger length 4.3 mm. This indicates that the carapace is actually slightly longer, not shorter, than the movable finger. Our measurements of the holotype (Table 1) indicate that these two structures are essentially equal in length. In addition, the ratio carapace length/movable finger length is quite variable in many species of Paruroctonus. Therefore, we do not consider this ratio to be a good diagnostic indica- tor of species differences. With respect to coloration, P. boreus has been characterized as being pale yellow to orange -brown with more or less developed dusky or black pattern on the carapace, mesosoma, and ventral aspect of the metasoma. P. aquilonalis has been characterized as SISSOM AND FRANCKE -PARUROCTONUS FROM NEW MEXICO AND TEXAS 95 Table 1. -Measurements (in millimeters) of Paruroctonus aquilonalis (Stahnke, 1940) [= P. boreus (Girard 1854)] , Paruroctonus williamsi, n. sp., and Paruroctonus pecos, n. sp. P. aquilonalis P. williamsi P. williamsi P. pecos P. pecos holotype <5 holotype 6 paratype 9 holotype 6 paratype 9 Total length 34.55 33.79 33.46 34.44 34.81 Carapace length 4.20 4.29 4.58 4.15 4.57 Mesosoma length 9.25 9.01 9.56 9.19 10.78 Metasoma length 21.10 15.94 14.57 16.15 14.80 I length/ width 2.20/2.15 2.10/2.15 1.73/2.25 2.10/2.25 1.98/2.40 II length/ width 2.60/2.05 2.43/2.03 2.30/2.05 2.53/2.13 2.30/2.18 III length/ width 2.90/1.95 2.71/1.91 2.50/1.92 2.93/2.02 2.49/2.08 IV length/ width 3.60/1.85 3.60/1.76 3.34/1.89 3.35/1.84 3.33/1.94 V length/ width 5.20/1.70 5.10/1.64 4.70/1.78 5.24/1.52 4.70/1.61 Telson length 4.60 4.55 4.75 4.95 4.66 vesicle width/depth 1.80/1.40 1.37/1.25 1.71/1.36 1.55/1.24 1.56/1.44 aculeus length 1.65 2.05 1.85 2.08 Pedipalp length 14.50 13.05 13.59 13.16 13.38 femur length 3.80 3.28 3.26 3.20 3.15 femur width 1.15 1.15 1.28 1.05 1.30 femur depth 1.20 1.25 1.00 1.05 tibia length 3.90 3.60 3.85 3.60 3.78 tibia width 1.40 1.51 1.74 1.69 1.69 tibia depth 1.45 1.54 1.47 1.41 chela length 6.80 6.17 6.48 6.36 6.45 chela width 1.90 2.29 2.09 2.19 2.13 chela depth 2.60 2.55 2.45 2.60 2.56 movable finger length 4.25 3.64 3.66 3.74 3.74 fixed finger length 3.30 2.63 2.71 2.57 2.69 Chelicera length 2.24 2.15 1.99 2.35 chela length/ width 1.35/1.09 1.35/1.25 1.15/1.05 1.50/1.20 movable finger length 1.29 1.50 1.20 1.40 fixed finger length 0.89 0.80 0.84 0.85 Pectinal tooth count 29-29 22-21 14-14 21-20 14-13 being uniformly orange-brown without variegations. First of all, the holotype of P. aqui- lonalis is in such poor state of preservation (it appears to have been dried on more than one occasion), that it is impossible to determine the original color and pattern (if any). Secondly, the supposed differences in color disappear when large samples are analyzed. We have examined several samples from northern Arizona and southern Utah with consid- erable variability in the development of dusky markings: a few specimens have a distinct P. boreus pattern, others are unvariegated as in P. aquilonalis , and the majority present a range of variability that spans the two extremes. Therefore, this character is unreliable and cannot be used to separate these two nominal taxa. Detailed comparisons of numerous samples from northern Arizona and southern Utah indicate that two closely related species of Paruroctonus inhabit the area. One, with variable color pattern, and with 6-7 pairs of setae along the ventrolateral keels of meta- somal segment V represents P. boreus. The other, always immaculate, and with 9-1 1 pairs of setae along the ventrolateral keels of segment V represents P. utahensis. The holotype of P. aquilonalis clearly belongs to the first species; therefore we propose the following synonymy: P. boreus (Girard, 1854) = P. aquilonalis (Stahnke, 1940). 96 THE JOURNAL OF ARACHNOLOGY Published records of P. boreus, under the name P. aquilonalis , from New Mexico, Texas, and Chihuahua (Gertsch and Soleglad 1966, Muma 1975, Diaz Najera 1975, Row- land and Reddel 1976, Riddle et al. 1976, Riddle and Pugash 1976, Riddle 1978) are all referable to P. utahensis. Paruroctonus utahensis (Williams, 1968) Figs. 1-6, 29-30 Type data.— Holotype male from 2 mi. NE Bluff, San Juan Co., Utah, 14 July 1967 (S. C. Williams, M. A. Cazier, J. Davidson). Deposited at the California Academy of Sciences, San Francisco. Distribution.— USA: Arizona, Utah, New Mexico, Texas. MEXICO: Chihuahua. For distribution in Texas and New Mexico, refer to Fig. 35. Diagnosis.— Adults 35-45 mm in length. Color pale yellow, always immaculate. Dorso- lateral carinae on metasomal segments I-IV well developed, serrate, strongly convergent posteriorly; setation 0: 1:1:2. Ventral submedian keels on I obsolete; on II-III poorly developed, smooth; on IV moderately developed, smooth to crenulate; setation typically 3:4:4:5. Ventrolateral keels on segment V bearing 9-11 pair of setae. Chelicerae with subdistal and distal teeth of movable finger subequal in length, apposed (Figs. 29-30); inferior margin of movable cheliceral finger with 4-5 weak crenulations. All carinae of pedipalp chela well developed, strongly granular; dorsal marginal and ventroexternal carinae with multiple rows of granules (Figs. 1-6). Pectinal tooth count 29-37 in males, 17-22 in females. Comparisons.— Based on general appearance and setation, P. utahensis is most similar to P. auratus (Gertsch and Soleglad) and P. boreus. It can be distinguished from P. auratus by the following characters: (1) P. utahensis has 9-11 pair of setae on the ventrolateral carinae of metasomal segment V, while P. auratus has 6 pair; (2) pectinal tooth counts for males of P. utahensis , although ranging from 29-37 , are typically 31-34, while the counts for males of P. auratus are lower, being 25-29 (Pectinal tooth counts in females of the two species are quite similar); and (3 )P. utahensis always lacks nodules at the base of the fixed cheliceral finger, which are present in P. auratus (2 nodules in males, 3 in females). P. utahensis differs from P. boreus in the following ways: (1) P. boreus has only six setal pairs along the ventrolateral carinae of metasomal segment V; and (2) the ventro- lateral carinae of segments II-IV are crenulate to serrate inP. utahensis , but smooth inP. boreus. Specimens examined (New Mexico and Texas records only). -NEW MEXICO: Bernalillo County. Tijueras Arroyo, just S Albuquerque, 2 June 1974 (W. A. Riddle), 7 99 (OFF), 14 June 1974, 15 dd (OFF); Albuquerque (4945 ft.), 5 Sept. 1971 (K. J. Teater), 1 d (OFF); 5 mi. N Interstate Highway 40, Albuquerque, 12 June 1969 (Cazier et al.), 7 dd, 11 99, 7 imm. (OFF): Chaves County. Mescalero Sands, 9 mi. W Caprock, 21 March 1980 (W. D. Sissom), 1 d, 7 99,4 imm. (WDS); 21 March 1980 (J. C. and J. E. Cokendolpher), 4 99, 5 imm. (JCC); 22 March 1980 (J. C. and J. E. Cokendolpher), 3 99, 5 imm. (JCC): Dona Ana County: Las Cruces (in city), 28 May 1970 (R. L. Smith), 2 imm. (OFF): Eddy County: 15 mi. E Loving, July 1978 (C. Rudolph) (oak shinnery), 13 dd, 1 imm., (creosote scrub and dunes) 8 dd, (open dunes) 30 dd, 4 99 (OFF): Lincoln County: Coyote (7000 ft.), no date (G. Vensel), 1 d (OFF): Otero County: White Sands National Monument (on consolidated sand at base of dunes), 25 July 1971 (S. Szerlip), 1 d, 2 99 (OFF): Valencia County: 20.4 mi. NW Los Lunas on New Mexico Highway 6, 17 June 1970 (M. A. Cazier, L. Welch, O. Francke), 10 dd, 11 99, 12 imm. (OFF). TEXAS: El Paso County: Anthony (on New Mexico boundary of Doha Ana Co.), 19 June 1970 (M. A. Cazier, L. Welch, O. Francke), 263 specimens (OFF), 24 June 1970, over 250 specimens (OFF); 4 June 1974 (M. A. Cazier, L. Draper, O. Francke), 40 dd, 69 99, 146 imm. (OFF): Ward County: Monahans Sand Hills State Park, 8 Sept. 1979 (R. Stewart et al.), 9 dd, 4 99 (OFF): Winkler County: 7.9 mi. NE Kermit, 6 April 1979 (W. D. Sissom), 2 99, 1 imm. (WDS). SISSOM AND FRANCKE -PARUROCTONUS FROM NEW MEXICO AND TEXAS 97 Paruroc tonus gracilior (Hoffmann , 1931) Figs. 7-12,33-34 Type data.— Male lectotype from Tepezala, Aguascalientes, Mexico (C. C. Hoffmann). Deposited at the American Museum of Natural History, New York; examined. Distribution.-USA: Arizona, New Mexico, Texas. MEXICO: Coahuila, Aguascalientes. For distribution in Texas and New Mexico, refer to Fig. 35. Figs. 1-3.- Right pedipalp chela of male Paruroctonus utahensis (Williams) from Eddy County, New Mexico, showing tricobothrial pattern: 1, dorsal aspect; 2, external aspect; 3, ventral aspect. Figs. 4-6. -Right tibia of male Paruroctonus utahensis (Williams) from Eddy County, New Mexico, showmg tricobothrial pattern: 4, dorsal aspect; 5 , external aspect; 6, ventral aspect. Scale = 1 .0 mm. 98 THE JOURNAL OF ARACHNOLOGY Diagnosis.— Adults 35-45 mm in length. Base color yellow to yellow-brown; interocular area of carapace marked by a dusky triangle; tergites with considerable dusky coloration. Dorsolateral carinae on metasomal segments I-IV well developed, serrate, weakly conver- gent posteriorly; setation 1 : 1 : 1 : 2. Ventral submedian keels on I obsolete; on II-III poorly developed, smooth; on IV moderately developed, smooth; setation 4:5:5:6-7. Ventro- lateral carinae on segment V bearing 9-11 pair of setae. Cheliceral movable finger with smaller subdistal tooth not in apposition with distal tooth (Figs. 33-34); distal tooth of movable finger greatly enlarged, strongly curved inward; inferior margin of movable finger usually with 2 or 3 large denticles, and a varying number of smaller ones. All carinae of pedipalp chelae well developed, granular (Figs. 7-12). Pectinal tooth count 24-29 in males, 18-23 in females. Comparisons.— See comparisons under P. williamsi. Specimens examined (New Mexico and Texas records only). -NEW MEXICO: Chaves County: 19.2 mi. W Caprock, 21 March 1980 (J. E. Cokendolpher), 1 9 (JCC): Eddy County: Carlsbad Caverns National Monument, 8 Sept. 1969 (M. A. Cazier, J. Bigelow), 4 dd, 5 99 (OFF); 15 mi. E Loving (creosote scrub and dunes), July 1978 (C. Rudolph), 9 dd (OFF): Hidalgo County: 13 mi. N Rodeo, 25 June 1973 (O. F. Francke), 2 imm. (OFF): Luna County: Rock Hound State Park, 9 mi. E Deming (on sand), no date (J. Bigelow), 1 d, 1 9 (OFF). TEXAS: Brewster County: Castolon, Big Bend National Park, 8 August 1979 (O. F. Francke, J. V. Moody), 1 d (OFF); Grapevine Ranch, N base of Grapevine Mountain, Big Bend National Park, 7 Sept. 1969 (M. A. Cazier, J. Bigelow), 15 dd, 3 99, 2 imm. (OFF): Culberson County: 1 mi. N Kent, 23 June 1970 (M. A. Cazier, L. Welch, O. F. Francke), 4 99, 3 imm. (OFF): Jeff Davis County: Phantom Spring, 25 June 1968 (J. C. Lewis), 1 9 (OFF): Presidio County: 3.3 mi. N Presidio, 2 Sept. 1972 (J. Davidson, O. F. Francke), 8 dd, 11 99, 3 imm. (OFF); 2.9 mi. E Presidio, 2 Sept. 1972 (J. Davidson, O. F. Francke), 6 dd, 3 99, 1 imm. (OFF); 36 mi. S Marfa, 2 May 1980 (L. Robbins), 2 imm. (OFF): Val Verde County: 0.5 mi. S Langtry, 14 June 1974 (L. Draper, M. A. Cazier, O. F. Francke), 1 9 (OFF): Winkler County: 3 mi. W Wink, 7 April 1979 (W. D. Sissom), 2 dd (WDS). Paruroctonus williamsi, new species Figs. 13-20,31-32 Type data.— Adult male holotype from Grapevine Ranch, north base of Grapevine Mountain, Big Bend National Park, Brewster Co., Texas, 7 September 1969 (M. A. Cazier and J. Bigelow). Deposited at the American Museum of Natural History, New York. Paratypes listed under specimens examined. Etymology.— The specific name is a patronym honoring Dr. Stanley C. Williams for his contribution to scorpion systematics. Distribution.— Known only from Big Bend National Park, Brewster Co., Texas (Fig. 35). Diagnosis.— Adults 30-40 mm in length. Base color yellowish brown. Carapace, includ- ing interocular area, and tergites with dusky pattern. Dorsal lateral carinae on metasomal segments I-IV strongly serrate, setation 1:3:3:3. Ventral submedian keels on I and II poorly developed, smooth; on III well developed, smooth; on IV well developed, smooth to crenulate; setation 3:4-5:4-5:5-7. Ventrolateral keels on segment V bearing 9-11 pair of setae. Chelicerae similar to P. gracilior, with smaller subdistal tooth not in apposition with distal tooth; inferior margin of movable cheliceral finger with 7-8 weak crenulations. Carinae of pedipalp chela poorly to moderately developed, smooth to granular. Pectinal tooth count 19-23 in males, 14 in females. Description.— The following description is based on males; parenthetical statements refer to females. Measurements of holotype and female paratype given in Table 1. SISSOM AND FRANCKE -PARUROCTONUS FROM NEW MEXICO AND TEXAS 99 Prosoma. Carapace: Base color yellow brown, mottled with dusky pattern; interocular area same color as adjoining carapace; anterior margin with slight emargination(straight), with eight setae; three pairs lateral eyes; ocular tubercle black, smooth. Anterior median furrow narrow, shallow; central transverse furrow shallow, wide; posterior median furrow deep; posterior lateral furrow shallow, wide, weakly arcuate. Anterior one-half of cara- pace rather smooth with few coarse granules, posterior one-half with numerous large granules (few, scattered) on areas surrounding furrows. Sternum subpentagonal with deep posteromedial depression; posterior margin strongly notched; slightly wider than long. Figs. 7-9.- Right pedipalp chela of male Paruroctonus gracilior (Hoffmann) from Eddy County, New Mexico, showing tricobothrial pattern: 7, dorsal aspect; 8, external aspect; 9, ventral aspect. Figs. 10-1 2. -Right tibia of male Paruroctonus gracilior (Hoffmann) from Eddy County, New Mexico, showing tricobothrial pattern: 10, dorsal aspect; 11, external aspect; 12, ventral aspect. Scale = 1.0 mm. 100 THE JOURNAL OF ARACHNOLOGY Table 2. -Variability in setation of the ventral submedian keels of segments I-IV in P. mlliamsi and P. pecos. P. williamsi P. pecos segments segments No. Setae I II III IV I II III IV 2-3 2 3-3 10 4 3-4 3 3 4-4 7 4 2 10 6 4-5 5 5 1 3 5-5 1 3 4 1 3 5-6 1 2 1 8 6-6 1 6-7 2 7-7 4 Mesosoma. Base color yellow brown, dusky pattern variable; yellow median stripe sometimes present along length of mesosoma. Posterior edges of tergites I-VI granular (smooth), anterior edges smooth. Median keel on I vestigial, on II represented by a few coarse granules (smooth), on III-VI by a row of coarse granules (smooth); VII tetra- carinate, all carinae serrate (crenulate), lateral margin with serrate (crenulate) keel. Ster- nites yellow brown, III-VI smooth, VII with one pair of smooth (obsolete) lateral keels. Genital operculum: Little more than 1.5 times as broad as long, posterior one-half com- pletely divided longitudinally; genital papillae present (absent). Pectines: Basal piece little more than 1.5 times as wide as long, with deep anteromedian notch; middle lamellae ovate to round, setate, numbering 15-20 (13-14); pectinal tooth count 19-23 (14). Metasoma. Yellow, venter with narrow dusky stripes underlying ventral submedian and ventrolateral carinae. Dorsolateral keels well developed, strongly serrate (weakly serrate), moderately convergent posteriorly, with 1:3:3:3 pairs of setae respectively. Lateral supra- median keels well developed, serrate to crenulate (as in male, but less pronounced), on I weakly convergent posteriorly. Lateral inframedian keels on I complete, crenulate; repre- sented on posterior one-fifth of II-III by four large granules; absent on IV. Ventrolateral keels on I-III well developed, smooth; on IV well developed, smooth to crenulate. Ventral submedian keels on I-II poorly developed, smooth; on III well developed, smooth; on IV well developed, smooth to crenulate. Setae of ventral submedian keels 3: 4-5: 4-5: 5-7. All intercarinal spaces shagreened. Segment V: Dorsolateral keels moderately developed, crenulate (smooth); lateral submedian keels moderately developed, incomplete, crenulate (smooth); ventrolateral keels well developed, strongly serrate, with 9-11 pair of setae; ventral median keel well developed, strongly serrate. Intercarinal spaces of ventral aspect with a few large granules, others shagreened. Telson. Vesicle yellow, flattened dorsally, 2.0 (1.5) times as long as wide, about as deep as wide (wider than deep), with 10 pairs of major setae. Aculeus reddish brown, moderately curved, about two-thirds as long as vesicle. Chelicerae (Figs. 31-32). Creamy yellow, teeth brown. Dentition similar to P. stahnkei (Gertsch and Soleglad): Distal and subdistal teeth of movable finger not in apposition, subdistal tooth not quite one-third the length of distal tooth. Inferior margin of movable finger with 7-8 weak crenulations, one or two of these usually larger than others. SISSOM AND FRANCKE-iMZ? UROCTONUS FROM NEW MEXICO AND TEXAS 101 Pedipalps. Femur: Base color yellow brown, intercarinal spaces dusky (plain). Dorso- internal keel well developed, granose; ventrointernal keel well developed, weakly serrate (granose); dorsoexternal keel well developed, crenulate (smooth to crenulate); ventro- external keel represented by three strong granules (smooth). Intercarinal spaces sha- greened. Orthobothriotaxia “c” (Vachon 1974). Tibia (Figs. 18-20): Base color yellow brown, intercarinal spaces dusky. Dorsointernal keel well developed, granular; basal tubercle moderately strong; ventrointernal keel well developed, serrate; dorsoexternal keel moderately developed, crenulate; external face of tibia covered with small granules; ventroexternal keel well developed, crenulate. Inter- carinal spaces shagreened. Orthobothriotaxia “C” (Vachon, 1974). Chela (Figs. 13-17): Base color yellow brown, dentate margins of chela fingers brown to reddish brown; fixed finger with six rows of granules flanked by six inner accessory granules, movable finger with six rows of granules flanked by seven inner accessory Figs. 13-17. -Right pedipalp chela of male holotype Paruroctonus williamsi, new species, showing tricobothrial pattern: 13, dorsal view; 14, dentate margin of movable finger; 15, external view; 16, ventral view; 17, dentate margin of fixed finger. Figs. 18-20.-Right tibia of male holotype Paruroctonus williamsi, new species, showing tricobo- thrial pattern: 18, dorsal view; 19, external view; 20, ventral view. Scale =1.0 mm. 102 THE JOURNAL OF ARACHNOLOGY granules. Dorsal marginal keel moderately developed, granular to smooth (smooth); dorsal secondary keel moderately developed, smooth; digital keel well developed, smooth; ex- ternal secondary keel poorly developed, granular to smooth; ventroexternal keel well developed, crenulate to serrate (smooth to crenulate); ventromedian keel moderately developed, crenulate (smooth); ventrointernal keel poorly developed, smooth; dorso- internal keel moderately developed, smooth to granular (smooth). Ratio of chela length to chela width approximately 2.70 in males, 3.10 in females; of fixed finger length to chela length approximately 0.40; of movable finger length to fixed finger length approxi- mately 1.35. Legs. Yellow, with or without pattern of dusky coloration. Tarsomere I of legs I-III with dorsal row of six to eight stout setae. Comparisons. -Based on cheliceral dentition, P. williamsi is most similar to P. graci- lior, P. stahnkei , and P. pallidus. FromP. gracilior, it differs in the following ways: (1) P. williamsi has 1 :3: 3:3 pair of setae on metasomal segments I-IV respectively, while P. gracilior has 1: 1:1:2 pair; (2) P. williamsi has 19-23 pectinal teeth in males (14 in females), and P. gracilior has 24-29 in males (18-23 in females); and (3) in P. williamsi , the distal tooth of the cheliceral movable finger is shorter and less curved. P. williamsi differs from P. stahnkei in the following ways: (1) P. williamsi has 9-11 pair of setae along the ventrolateral carinae of segment V, while P. stahnkei has only 6-7 pair; (2) P. stahnkei has typically 0: 1:1:2 pair of setae on the dorsolateral carinae of segments I-IV; and (3) P. williamsi has few granules on the ventral surface of metasomal segment V, while in P. stahnkei the granules are numerous. P. williamsi differs from P. pallidus in the following ways: (1) P. pallidus has 0: 1 : 1 :2 pair of setae on the dorsolateral carinae of segments I-IV; (2) P. pallidus has more than 24 pectinal teeth in males (more than 17 in females); and (3) P. pallidus has only 8 pair of setae along the ventrolateral carinae of segment V. Variation.— Adults of P. williamsi vary slightly in the amount of dusky coloration on the legs, pedipalps, and cauda. Several specimens possess a rather distinct median yellow stripe along the length of the mesosoma. The number of pectinal teeth ranges from 19 to 23 in males. In specimens examined, there are three combs with 19 teeth, nine combs with 20 teeth, eight combs with 21 teeth, two combs with 22 teeth, and one comb with 23 teeth. A comb of one male and combs of two other males were damaged and could not be used in the counts. The single female examined has 14 teeth on each comb. Considerable variation occurs in the setation of the ventral submedian keels of meta- somal segments I-IV. The setae of these keels of this (and many other) species of Paruroc- tonus are paired, and variation not only occurs in the number of setal pairs, but there may also be extra setae in one or both rows. In other cases setal rows are staggered so that pairing cannot be discerned. Therefore, setation formulae should be used very carefully when identifying specimens, and key couplets should not be heavily based on this char- acter (e.g., Soleglad 1972). To recognize the variability in this character, the setal formu- lae have been constructed so that the segments are separated by a colon, and variation in the number of pairs of individual segments by a hyphen. Only common variations of the latter are used in the formula. The setal formula for the ventral submedian keels in P, williamsi is 3:4-5:4-5:5-7, showing a high degree of variability in this character for this species. For a more complete analysis consult Table 2. Specimens examined. -TEX AS: Brewster Co.', Grapevine Ranch, north base of Grapevine Moun- tain, Big Bend National Park, 7 September 1969 (M.A. Cazier and J. Bigelow), holotype <5, 1 9 paratopotype (AMNH), 13 dd paratopotypes (OFF), 2 dd paratopotypes (WDS). SISSOM AND FRANCKE -PARUROCTONUS FROM NEW MEXICO AND TEXAS 103 Paruroctonus pecos , new species Figs. 21-28 Type data.-Adult male holotype from 15 mi. E Loving Eddy Co., New Mexico, July 1978 (C. Rudolph). Deposited at the American Museum of Natural History, New York. Paratypes listed under specimens examined. Etymology. -The specific name is a noun in apposition taken from the Pecos River, which is near the type locality. Distribution.— Known only from the type locality and the Mescalero Sands Region, Chaves Co., New Mexico (Fig. 35). Diagnosis.— Length 30-40 mm. Base color yellow brown. Carapace with dusky cres- cent delineating interocular triangle; tergites with dusky pattern. Dorsolateral carinae on metasomal segment I moderately developed, crenulate; on II-IV well developed, serrate; setation 1:1: 1:2. Ventral submedian keels on I very poorly developed, smooth to obso- lete; on II-III poorly developed, smooth to obsolete; on IV moderately developed, smooth to crenulate; setation 3-4:4:4-5:5-6. Ventral surface of metasomal segment V with few scattered granules. Chelicerae similar to P. stahnkei, with smaller subdistal tooth not in apposition with distal tooth; inferior margin of movable cheliceral finger with seven to eight small denticles. Carinae of pedipalp chela moderately to well developed, smooth to granular. Pectinal tooth count 20-22 in males, 13-15 in females. Description.— The following description is based on males; parenthetical statements refer to females. Measurements of holotype and paratype female given in Table 1 . Prosoma. Carapace: Base color yellow brown; interocular area marked by a crescent shaped area of dusky coloration; posterior portions of carapace dusky; anterior margin essentially straight, with eight setae; three pair lateral eyes; ocular tubercle black, smooth. Anterior median furrow narrow, shallow; central transverse furrow shallow, wide; poste- rior median furrow moderate to deep; posterior lateral furrow moderately deep, wide, weakly arcuate. Entire carapace covered with evenly spaced granules. Sternum subpenta- gonal with deep posteromedial depression; posterior margin strongly notched; about as long as wide. Mesosoma. Base color yellow brown, dusky pattern variable; median yellow stripe present along length of mesosoma; surface shagreened. Tergites I-II smooth; posterior edges of III- VI with few (small) scattered granules, anterior edges smooth. Median keel vestigial on I, obsolete on II-VI. Tergite VII tetracarinate, all carinae serrate (crenulate to serrate); lateral margin of VII with crenulate keel. Sternites yellow to yellow brown, III- VI smooth, lustrous; VII with one pair of vestigial keels. Stigmata elongate oval. Genital operculum: Little more than twice as broad as long; completely divided longitu- dinally; genital papillae present (absent). Pectines: Basal piece not quite 1.5 times as broad as long, with deep anteromedian notch; middle lamellae ovate to round, numbering 15-17 (10-13); pectinal tooth count 20-22 (13-15). Metasoma. Yellow brown, with variable amounts of dusky coloration. Segments I-IV: Dorsolateral carinae on I moderately developed, crenulate, moderately convergent poste- riorly; on II-IV well developed, serrate, moderately to weakly convergent posteriorly; setae on dorsolateral carinae 1:1:1 :2- Lateral supramedian keels on I-IV moderately to well developed, crenulate. Lateral inframedian keels on I complete, crenulate; represented on posterior one-fifth of II-III by two to three granules; absent on IV. Ventrolateral keels on I moderately developed, smooth; on II-III well developed, smooth; on IV well devel- oped, smooth to crenulate. Ventral submedian keels on I very poorly developed, absent 104 THE JOURNAL OF ARACHNOLOGY on anterior one-half, smooth on posterior one-half; on II-III poorly developed, smooth; on IV moderately developed, smooth to crenulate; setae on ventral submedian keels 3-4:4:4-5:5-6. All intercarinal spaces shagreened. Segment V: Dorsolateral keels moder- ately developed, smooth; lateral submedian keels moderately developed, granular, incom- plete; ventrolateral keels well developed, serrate, with 9-10 pair of setae; ventromedian keel well developed, serrate. Intercarinal spaces of ventral aspects of V with a few small granules, all other intercarinal spaces shagreened. Telson. Vesicle yellow brown, flattened dorsally; twice (1.5 times) as long as wide, little wider than deep, with 1 1 pairs of major setae; subtle subaculear tubercle sometimes present in males. Aculeus reddish brown, moderately curved, about five-eighths as long as vesicle. Figs. 21-23. -Right pedipalp chela of male holotype Paruroctonus pecos, new species, showing trichobothiial pattern: 21 , dorsal view; 22, external view; 23, ventral view. Figs. 24-26. -Right tibia of male holotype Paruroctonus pecos, new species, showing tricobothrial pattern: 24, dorsal view; 25, external view; 26, ventral view. Scale =1.0 mm. SISSOM AND FRANCRE-PARUROCTONUS FROM NEW MEXICO AND TEXAS 105 Figs. 27-34.-Right chelicerae of species of Paruroctonus in New Mexico and Texas: Figs. 27, 29, 31, 33.-Dorsal views; Figs. 28, 30, 32, 34. -Prolateral views of movable finger; Figs. 27-28: Paruroc- tonus pecos, holotype male; Figs. 29-30: Paruroctonus utahensis, male from Eddy County, New Mexico; Figs. 31-32: Paruroctonus williamsi, holotype male; Figs. 33-34 : Paruroctonus gracilior , male from Eddy County, New Mexico. Scale =1.0 mm. 106 THE JOURNAL OF ARACHNOLOGY Chelicerae (Figs. 27-28). Yellow, teeth brown. Dentition similar to P. stahnkei and P. williamsi : Distal and subdistal teeth of movable finger not in apposition, subdistal tooth slightly more than one-third length of distal tooth. Inferior margin of movable cheliceral finger with 7-8 small denticles, two of these usually smaller than others. Pedipalps. Femur yellow brown, with variable dusky pattern. Dorsointernal keel well developed, crenulate (granular); ventrointernal keel well developed, crenulate to serrate (crenulate); dorsoexternal keel moderately developed, granular (smooth); ventroexternal keel marked by three to four granules. Internal face with numerous small granules, other faces shagreened. Orthobothriotaxia “C” (Vachon 1974). Tibia (Figs. 24-26). Color yellow brown, with variable dusky pattern. Dorsointernal keel well developed, serrate (smooth to crenulate); basal tubercle strong; ventrointernal keel well developed, crenulate to serrate (crenulate); dorsoexternal keel well developed, smooth; external face of tibia covered with small granules; ventroexternal keel well devel- oped, granular. Intercarinal spaces shagreened. Orthobothriotaxia “C” (Vachon 1974). Chela (Figs. 21-23). Color yellow brown, lustrous; dentate margins of chela fingers brown to reddish brown, with six rows of granules flanked by six inner accessory granules on fixed finger, and six rows of granules flanked by seven inner accessory granules on movable finger. Dorsal marginal keel moderately (poorly) developed, granular (smooth); dorsal secondary and digital keels moderately developed, smooth; external secondary keel Fig. 35. -Map showing distribution of the genus Paruroctonus in Texas and New Mexico. SISSOM AND FRANCKE-iMR UROCTONUS FROM NEW MEXICO AND TEXAS 107 poorly developed, smooth; ventroexternal keel well developed, granular to crenulate (smooth to crenulate); ventromedian keel moderately developed, smooth; ventrointernal keel poorly developed, smooth; dorsointernal keel moderately developed, smooth to granular (smooth). Ratio of chela length to width about 2.90; of fixed finger length to chela length about 0.40; of movable finger length to fixed finger length 1 .45. Legs. Base color yellow; faint dusky coloration present on femora and patellae. Tarso- mere I of legs I-III with dorsal row of 6-8 stout setae. Comparisons.— Based on cheliceral dentition and carinal development, P. pecos is most similar to P. williamsi and P. stahnkei . From P. williamsi it differs in the following ways: (1) P. pecos has 1:1:1 :2 pair of setae on the dorsolateral carinae of metasomal segments I-IV, P. williamsi has 1:3: 3: 3 pair; (2) P. pecos has the interocular area of the carapace delineated by a dusky crescent, whereas P. williamsi lacks this; and (3) in P. pecos metasomal carinal development is much weaker than in P. williamsi. Paruroctonus pecos differs from P. stahnkei in the following ways: (1)R. pecos has 9-10 pair of setae along the ventrolateral carinae of metasomal segment V, whereas P. stahnkei has only 6-7 pair; (2) in P. pecos the granules on the ventral surface of meta- somal segment V are few and sparsely distributed, but in P. stahnkei they are numerous and rather densely distributed; and (3) P. pecos has 1 : 1 : 1 : 2 pair of setae on the dorso- lateral carinae of segments I-IV, and P. stahnkei typically has 0: 1 : 1 :2. Variation.— Adults of P. pecos vary in the amount of dusky coloration on the carapace, pedipalps, legs, and cauda. Juveniles generally have more dusky markings and are paler in color than adults. Pectinal tooth counts vary in males examined from 20-22. On males there are two combs with 20 teeth, four combs with 21, and two combs with 22. Pectinal tooth counts in females vary from 13-15. On females there are three combs with 13 teeth, nine combs with 14, and two combs with 1 5. The setal formula for the ventral submedian carinae of metasomal segments I-IV is 34:4:4-5:5-6. For a more detailed analysis of the setal variation, consult Table 2. The ventrolateral carinae of metasomal segment V vary from 9-10 pair (one specimen has 8 pair). Remarks.— Specimens of Paruroctonus pecos collected in Chaves Co., New Mexico were found along the roadside of Highway 380 in the Mescalero Sands area. The sand is consolidated, with a relatively dense cover of buffalo grass, yucca, and creosote scrub. A single specimen of P. gracilior was collected under the same conditions. However, on the dunes where sand is unstable and vegetation sparse, P. utahensis was the only species collected. Specimens examined. -NEW MEXICO: Chaves County : 19.2 mi. W Caprock (on roadside), 21 March 1980 (W. D. Sissom), 1 imm. 6 (WDS); 20.5 mi. W Caprock (on roadside), 22 March 1980 (W. D. Sissom), 4 99, 1 imm. <5, 1 imm. 9 (WDS), 2 99 (OFF): Eddy County : 15 mi. E Loving, July 1978 (C. Rudolph), holotype <5 (AMNH), 1 6 paratopotype (OFF). ACKNOWLEDGMENTS We wish to thank Dr. N. I. Platnick, American Museum of Natural History, for allow- ing us to examine the types of Paruroctonus gracilior , and Dr. H. L. Stahnke, Tempe, Arizona, for allowing us to study the holotype of Paruroctonus aquilonalis. We also thank Dr. C. S. Crawford, University of New Mexico, for sending specimens collected in New Mexico. Specimens studied are deposited in the following collections: American Museum 108 THE JOURNAL OF ARACHNOLOGY of Natural History (AMNH), O. F. Francke (OFF), W. David Sissom (WDS), and J. C. Cokendolpher (JCC). This study was partially supported by The Institute for Museum Research, Texas Tech University. LITERATURE CITED Bugbee, R. E. 1942. Notes on animal occurrence and activity in the White Sands National Monument, New Mexico. Trans. Kansas Acad. Sci., 45:315-321. Diaz-Najera, A. 1975. Listas y datos de distribution geografica de los alacranes de Mexico. Rev. Inv. Salud Pub., Mexico, 35:1-36. Gertsch, W. J. and M. E. Soleglad. 1966. The scorpions of the Vejovis boreus Group (subgenus Paruroctonus ) in North America. American Mus. Novit., 2278:1-54. Girard, C. 1854. Arachnidians. In Marcy, R. B., Exploration of the Red River of Louisiana in the year 1852. Washington, pp. 251-261. Hoffmann, C. C. 1931. Los Scorpiones de Mexico. Primera Parte: Diplocentridae, Chactidae, Vejovidae. An. Inst. Biol., Mexico, 2:291-408. Muma, M. H. 1975. Two vernal ground-surface arachnid populations in Tularosa Basin, New Mexico. Southwestern Nat., 20:55-67. Riddle, W. A. 1978. Respiratory physiology of the desert grassland scorpion Paruroctonus utahensis. J. Arid Environ., 1:243-251. Riddle, W. A., C. S. Crawford, and A. M. Zeitone. 1976. Patterns of hemolymph osmoregulation in three desert arthropods. J. Comp. Physiol., 112:295-305. Riddle, W. A. and S. Pugach. 1976. Cold hardiness in the scorpion, Paruroctonus aquilonalis. Cryo- biology, 13(2): 248-25 3. Rowland, J. M. and J. R. Reddell. 1976. Annotated checklist of the arachnid fauna of Texas (exclud- ing Acarida and Araneida). Occas. Papers Mus., Texas Tech Univ., 38:1-25. Soleglad, M. E. 1972. Two new scorpions of the genus Paruroctonus from southern California. Was- mann J. Biol., 30:71-86. Stahnke, H. L. 1939. The scorpions of Arizona. Ph. D. Dissertation. Iowa State Univ., Ames. 185 pp. Stahnke, H. L. 1940. The scorpions of Arizona. Iowa St. College J. Sci., 15(1):101-103. Vachon, M. 1974. Etude des caracteres utilises pour classer les families et les genres de scorpions (Arachnides.) 1. La trichobothriotaxie en Arachnologie. Sigles trichobothriaux et types de tricho- bothriotaxie chez les scorpions. Bull. Mus. Nat. Hist. Nat., Paris, ser. 3, 140 (Zool. 104):857-958. Werner, F. 1934. Scorpiones, Pedipalpi. In Bronn, H. G. (ed.), Klassen und Ordnungen des Tierreich. Leipzig, vol. 5, pt. 4, book 8, pp. 1-316. Williams, S. C. 1968. Two new scorpions from western North America. Pan-Pacific Entomol., 44:313-321. Manuscript received April 1980, revised June 1980. The Journal of Arachnology 9:109 RESEARCH NOTES REPORT OF NECROPHAGY IN THE BLACK WIDOW SPIDER, LA TRODECTUS HESPER US ( ARANEAE: THERIDIIDAE) An unusual feeding behavior was observed in the western American black widow spider, Latrodectus hesperus Chamberlin and Ivie. One mature and one immature (sixth instar) female were introduced into a fifteen liter glass observation cage previously occupied by another spider. Both individuals actively explored the container. When the adult’s legs made contact with a dead fly (Syrphidae) on the substrate, the spider immediately began throwing silk about it. This is one of the initial steps in the stereo- typed prey capture sequence (Ross 1979). Both spiders were later observed feeding independently on the carcass. A second adult female, similarly placed in the abandoned container, was later found feeding on a large acridid presumed to be dead for over four days. These observations are most unusual in that black widows and all other spiders are typically regarded as being strictly predacious (Bristowe 1958). Exceptions to the rule of total predation in the Arachnida are known to occur among members of the orders Acari, Opiliones, and Solifugae (Savory 1977). Necrophagy in the black widow may serve as an accessory protein acquisition mechanism for use in emergencies. If this behavior is of more than occasional occurrence, it may supplement the ability of these spiders to fast for extended periods (Kaston 1970), enabling them to survive times of prey scarcity. REFERENCES CITED Bristowe, W. S. 1958. The World of Spiders. Collins Press, London. 304 pp Kaston, B. J. 1970. Comparative biology of American black widow spiders. Trans. San Diego Soc. Nat. Hist., 16: 33-82. Ross, K. G. 1979. Aspects of the biology of the black widow spider, Latrodectus hesperus Chamberlin and Ivie (Araneae, Theridiidae). Unpublished M. Sc. Thesis. University of Arizona, Tucson. Savory, T. H. 1977. Arachnida. Academic Press, London. 340 pp. Kenneth G. Ross, Department of Entomology Cornell University Ithaca, NY 14853. Manuscript received September 1979, revised October 1979. The Journal of Arachnology 9:110 REDESCRIPTION OF CHTHONIUS VIRGIN ICUS CHAMBERLIN (PSEUDOSCORPIONIDA, CHTHONIIDAE) Recent papers on Globochthonius Beier, a subgenus of Chthonius G. Koch, (Curcic 1973, 1976, 1977) have brought to mind the problem of Chthonius virginicus Chamberlin (1929), a species placed in Globochthonius by Beier (1931). While most species of Globo- chthonius are found in Europe, in the southern Alps and the Balkan peninsula (Curcic 1976), C. virginicus is apparently distributed over a small area of the east coast of the United States (Hoff 1958). These distributions would pose an interesting biogeographical problem if the species were correctly placed. However, as is shown below, C. virginicus actually belongs in the subgenus Ephippio chthonius Beier (1930). I am grateful to Dr. L. L. Pechuman for allowing me to study the type specimens in the Cornell University Collection, and to Charlotte H. Alteri for drawing the figures. Chthonius (Ephippiochthonius) virginicus Chamberlin Figs. 1 and 2 Material.— Holotype female (JC 64.01003) and one paratype female from Great Falls, Fairfax County, Virginia, 3 April 1921 (Crosby) [Cornell University Collection] . Description.— All parts lightly sclerotized and light in color. Carapace about as long as broad; no epistome, but anterior margin serrate across most of its width; four corneate eyes, the anterior pair larger than the posterior; chaetotaxy mm4mm-6-4-2-m2m, on each side two microsetae preocular and one laterally on the posterior margin. Coxal area typical; each coxa II with a group of eight bipinnate spines, and each coxa III with three similar spines; intercoxal tubercle with two small setae. Abdomen typical; tergal chaeto- taxy 4:4:4:4:6:6:6:6:6:4:6:0; sternal chaetotaxy 8:(3)10(3):(2)9(2): 10:7 :6:6:7:8:0:2, the lateralmost setae on sterna 5, 6 and 7 being very small. Figs. 1 and 2 .-Chthonius virginicus Chamberlin: 1, ventral view of left palp; 2, lateral view of right chela. The Journal of Arachnology 9:111 Chelicera 0.8 as long as carapace; hand with seven or eight setae ; movable finger with a row of five teeth, fixed finger with seven and an irregular ridge proximally ; flagellum of about 10 finely pinnate setae; galea a prominent knob. Palp characteristic of the subgenus (see Beier 1963:50); proportions of segments shown in Fig. 1; trochanter 1.55-1.6, femur 5.45-5.6, tibia 2.0-2. 1, and chela 4.4-4.65 times as long as broad; hand 2.05-2.2 times as long as deep; movable Finger 1.15 times as long as hand. Trichobothria as shown in Fig. 2, which also clearly illustrates the dorsal depression characteristic of Ephippiochthonius. Fixed chelal finger with 16 widely spaced teeth distally, followed by five very small denticles, and with a small accessory tooth on the external surface near the distal end; movable finger with six spaced teeth and six small proximal denticles; movable finger also with a small sensillum on the external surface proximal to the level of trichobothrium sb. Legs typical of the genus; leg IV with entire femur about 2.5 and tibia about 4.0 times as long as deep. Tactile setae not determinable. Measurements (mm).— Figures are given first for the holotype, followed in parentheses by those for the paratype. Body length 1.54(1.72). Carapace length 0.48(0.445). Chelicera 0.385(0.37) by 0.21(0.18). Palpal trochanter 0.18(0.17) by 0.11(0.11); femur 0.60(0.56) by 0.11(0.10); tibia 0.25(0.23) by 0.125(0.11); chela 0.835(0.79) by 0.19(0.17); hand 0.39(0.35) by 0.19(0.16); movable finger 0.445(0.415) long. Leg IV: entire femur 0.56(0.52) by 0.235(0.22); tibia 0.33(0.325) by 0.09(0.08); metatarsus 0.21(0.19) by 0.065(0.065); telotarsus 0.36(0.33) by 0.045(0.04). Remarks.— Inasmuch as no other members of Globochthonius have been identified from America, it is satisfying, zoogeographically speaking, to learn that C. virginicus does not belong to that subgenus, which, then, is apparently restricted to Europe. Problems remain, however, with regard to the identity and relations of C. virginicus with other members of the subgenus Ephippiochthonius , particularly C. tetrachelatus . Careful study is needed of the many American specimens that have been referred to C. tetrachelatus (see Hoff 1958) to determine whether they might belong to C. virginicus instead. Because the subgenus Ephippiochthonius is essentially European in distribution, it has been con- sidered that C. tetrachelatus had been introduced into America from Europe. Likewise, C. virginicus might be of European origin; if so, to what species does it correspond? An answer to this question can only be obtained after the status of the C. virginicus in America is ascertained and a more complete description is available, including the male and the nymphal stages. LITERATURE CITED Beier, M. 1930. Neue Hohlen-Pseudoscorpione der Gattung Chthonius. Eos (Madrid), 6: 323-327 . Beier, M. 1931. Zur Kenntnis der Chthoniiden (Pseudoskorpione). Zool. Anz., 93: 49-56. Beier, M. 1963. Ordnung Pseudoscorpionidea. Bestimmungsbiicher zur Bodenfauna Europas, 1 : 1-313. Chamberlin, J. C. 1929. A synoptic classification of the false scorpions or chela-spinners, with a report on a cosmopolitan collection of the same-Part I. The Heterosphyronida (Chthoniidae) (Arachnida- Chelonethida). Ann. Mag. Nat. Hist. ser. 10, 4: 50-80. Curcic, B. P. M. 1973. Le sous-genre Globochthonius Beier 1931 dans la Mediterranee nord- occidentale: Chthonius (G.) globifer Simon 1879 (Chthoniidae, Pseudoscorpiones, Arachnida). Wiss. Mitt. Bos.-herzegov. Landesmus, 3C: 77-84. Curcic, B. P. M. 1976. Le sous-genre Globochthonius Beier 1931 (Chthoniidae, Pseudoscorpiones): Considerations taxonomiques et implications biogeographiques. Bull. Acad. Serbe Sci. Arts, Cl. Sci. Math. Nat., 54(14): 21-27. The Journal of Arachnology 9:112 Curcic, B. P. M. 1977. Les voies de revolution morphologiques des pseudoscorpions mediterraneens. I. Le sous-genre Globochthonius Beier 1931 (Chthoniidae, Pseudoscorpiones, Arachnida). Proc. 6th Intern. Congr. Speleol., Olomouc 1973, 5: 47-49. Hoff, C. C. 1958. List of the pseudoscorpions of North America north of Mexico. Amer. Mus. Novitates, 1875: 1-50. William B. Muchmore, Department of Biology, University of Rochester, Rochester, New York 14627. Manuscript received September 1979, revised January 1980. STATUS OF LEIOBUNUM SERRA TIPALPE ROEWER (OPILIONES, LEIOBUNIDAE) The validity of the harvestman Leiobunum serratipalpe Roewer (1910) was first questioned by Crosby and Bishop (1924), who suggested that L. serratipalpe was pro- bably represented only by immatures of Leiobunum calcar (Wood, 1870). Davis (1934) reluctantly retained L. serratipalpe , but remarked on the similarity in the male genital morphology between L. serratipalpe and L. calcar. Roewer (1923, 1957), Bishop (1949), and Edgar (1966, 1971) retained L. serratipalpe and L. calcar as distinct species. My studies show that L. serratipalpe is not a distinct species but is a junior synonym of L. calcar. The original description of L. serratipalpe (Roewer 1910) was based on two specimens from Long Lake (New York ?) and Cold River (state unknown). Later, Roewer (1923, 1957) referred to these two as the “male type” and male “cotype.” I have been unable to examine the “male type,” presumably deposited in the Museum of Budapest, Hungary; but, from published descriptions (Roewer 1910, 1923) of the body size and palpal tarsi, ' it is probably a female. Through the kindness of Dr. M. Grasshoff I was able to examine the “male cotype” (cat. no. RII/32, and slide no. 10713) deposited in the Natur-Museum Senckenberg, Frankfurt am Main, West Germany. The “male cotype” is an adult female L. calcar. Davis (1934) and Bishop (1949) described and illustrated adult males which they referred to as L. serratipalpe. These males were characterized by having a small spur or group of denticles on the palpal femora and by having the penes relatively straight in lateral view (Davis 1934, figs. 32 and 3; Bishop 1949, figs. 81 and 83). Through the efforts of Drs. D. T. Jennings, Northeastern Forest Experiment Station, Orono, Maine, and M. W. Houseweart, Cooperative Forest Research Unit, University of Main, Orono, Maine, I was able to examine a large series of L. calcar collected in Piscataquis Co., Maine. The specimens were collected in large capacity pitfall traps (Houseweart et al. 1979) deployed in dense spruce-fir forests and in correspondingly similar forest harvested by strip cutting. Traps were installed on 26 May 1977, and removed 4 August 1977, for a 10-week trapping period; traps were installed again on 18 May 1978, and removed on 3 August 1978, for an 1 1-week trapping period. The Journal of Arachnology 9:113 Among more than 2000 specimens of L . calcar collected, there were 689 adult males, of which five had reduced palpal spurs. Of these five, only one had the genital mor- phology characteristic of L. serratipalpe\ the other four had the penes curved in lateral view (Davis 1934, fig. 16; Bishop 1949, fig. 48). Examination of specimens with palpal spurs showed both forms of penes as well as intermediates. No other differences could be found. LITERATURE CITED Bishop, S. C. 1949. The Phalangida (Opiliones) of New York. Proc. Rochester Acad. Sci., 9:159-235. Crosby, C. R. and S. C. Bishop. 1924. Notes on the Opiliones of the southeastern United States with descriptions of new species. J. Elisha Mitchell Sci. Soc., 40:8-26. Davis, N. W. 1934. A revision of the genus Leiobunum (Opiliones) of the United States. Amer. Midland Nat., 15:662-705. Edgar, A. L. 1966. Phalangida of the Great Lakes Region. Amer. Midland Nat., 75:347-366. Edgar, A. L. 1971. Studies on the biology and ecology of Michigan Phalangida (Opiliones). Misc. Pubs. Mus. Zool., Univ. Michigan, No. 144, 64 pp. Houseweart, M. W., D. T. Jennings and J. C. Rea. 1979. Large capacity pitfall trap. Ent. News, 90:51-54. Roewer, C. F. 1910. Revision der Opiliones Plagiostethi (=Opiliones Palpatores). 1. Teil: Familie der Phalangiidae (Subfamilien Gagrellini, Liobunini, Leptobunini). Abhandl. Naturwiss. Ver. Hamburg, 19(4): 1-294. Roewer, C. F. 1923. Die Weberknechte der Erde, Systematische Bearbietung der bisher bekannten Opiliones. Gustav Fisher, Jena, 1116 pp. Roewer, C. F. 1957. Uber Oligolophinae, Caddoinae, Sclerosomatinae, Leiobuninae, Neopilioninae und Leptobuninae (Phalangiidae, Opiliones, Palpatores). Senckenberg. Biol., 38:323-358. James C. Cokendolpher, The Museum and Department of Biological Sciences, Texas Tech University, Lubbock, Texas 79409. Manuscript received April 1980, revised May 1 980. The Journal of Arachnology 9:114 BOOK REVIEW LES ARAIGNEES: GENERALITES-ARAIGNEES DE FRANCE ET DES PAYS LIMI- TROPHES by Michel Hubert, Societe Nouvelle des Editions Boubee, Paris. 227 pp. 1979. (Price FF 150.—). Available in North America from Somabec Ltee, 2475 Sylva Clapin, C.P. 295, St-Hyacinthe, Quebec J2S 5T5. Michel Hubert is Maitre Assistant at the Museum d’Histoire Naturelle in Paris. He has a good working knowledge of the spiders of his country, and in this book he shares his expertise with readers who wish an introduction to the spiders in the French language. This book is similar in approach to Kaston’s “How to Know the Spiders,” with chapters on collecting and preserving (6 pages), anatomy (24 pages), and general biology (23 pages), plus a systematic part giving keys and brief descriptions for 33 families, 214 genera, and about one-fourth of the species of spiders found in France and its border countries (180 pages). The key to families, however, includes an additional 23 that are regarded as exotic and whose names appear in different print. The biological information is somewhat less than current and might have included recent advances such as Homann’s findings on eye structure. The nomenclature is that of Simon with only cosmetic improvements; the concept of haplogynes and entelegynes is perpetuated, as is that of salticid classification based on number of cheliceral teeth. Literature references are mainly to general works, not to revisions that could lead to more precise identifications. The book is illustrated with 16 colored plates and 228 (not 230 as claimed) line and stipple text figures. Eight of the plates depict, in habitus, 56 common European spiders by Jacques Rebiere, and also the see-through spider taken from Chapter II in Comstock’s “Spider Book;” the other plates are first rate photographs of spiders or their eggs by P. Lome and J. Six. The dust jacket on the book gives two additional color photos. The text figures are too few; 384 species are illustrated by 182 figures. Representatives of many genera are not illustrated at all, though described, and this will frustrate users of the book. Palpi of males are shown in one view only, and spermathecae are omitted. Is the book, as hopefully claimed in the preface by Professeur Deboutteville, of use to francophones outside Europe? A Quebecois entomological colleague assures me that at least the students of his province should welcome it, as approximately 60 percent of the genera treated in the book are also represented in North America. An introduction to the spiders in French has been a notable lack till now. Hubert has covered the ground thinly, perhaps aiming at a wide audience. This review- er feels that arachnology in France would have been better served by a family-by-family treatment of all the genera and species of spiders known to occur in the country— in short, an update of Simon’s “Arachnides de France.” Charles D. Dondale, Biosystematics Research Institute, Research Branch, Agriculture Canada, Ottawa, Ontario K1A OC6. The Journal of Arachnology 9:115 W.J. BAERG, 1885 - 1980 Dr. W. J. Baerg, Professor Emeritus of Entomology, University of Arkansas, died in Fayetteville, Arkansas on April 14, 1980. During his 31 years as Head of the Department of Entomology and for many years following his retirement in 1951, he devoted much of his energy and interest to the study of arachnids, especially tarantulas, scorpions, and species reported to be poisonous to man. Although his interests extended also to Ozark birds and flora as well as pest insects, 20 of his 75 published papers dealth with Arachnida, culminating with a small book, The Tarantula, a narrative account of his long personal association with those spiders. Coming to his first academic appointment at the University of Arkansas from Cornell in 1918, he saw his first tarantula— a tarantula ghetto, in fact, on a hillside on the university campus which has long since been displaced by fraternity houses-and his enchantment-at-first-sight with the creature remained all of his life. From about that time on he was in correspondence with arachnologists world-wide, and he was one of ten living American arachnologists listed in Bonnet’s Bibliographia Araneorum when it was publish- ed just after World War II. It was one of the pleasures of his later years that Bonnet visited him personally in 1975 after some 50 years of correspondence. Although Professor Baerg made extended observations on perhaps as many as 35 species of Central American, Mexican, Caribbean, and North American tarantulas, he became exasperated and finally disenchanted with the vagaries of mygalomorph taxon- omy. He was especially intrigued, however, with the apparent discrepancy between the reputation for toxicity and its actual effects, not only of the tarantulas but of other arachnids as well. He induced numbers of them often with difficulty, to bite or sting him and recorded his personal response to the effects of their venom. At the age of 85 he volunteered to be the experimental animal to test the toxicity on humans of Chiracan- thium inclusum (Hentz) whose biology I was at the time studying in his laboratory. (I declined his offer, and the effects of its bite are still uncertain.) Between 1924 and 1955 he made numbers of trips in connection with his studies, to Europe in 1929 and 1931, to Panama in 1924 and 1936, as a Fulbright Scholar to Jamaica in 1951 and 1952, and as a Guggenheim Fellow to Mexico in 1954 and 1955. He worked with Snodgrass at the National Museum on the internal anatomy of tarantulas and made numerous excursions throughout the United States and Mexico studying scor- pions and tarantulas in the field. Such was his devotion to the tarantula that he considered that all of his students of entomology should at least make its acquaintance. For some 30 years that he taught beginning entomology he would introduce the students to the large native species by having them pass one from hand to hand around the class. Only one person was ever bitten, he averred, and many a character was strengthened. Numbers of his former students have achieved national and international recognition in the field of entomology, and he inspired and endeared himself to many other entomo- logists and arachnologists, among whom I was one of the privileged. Although our The Journal of Arachnology 9:116 connections were intermittent, they spanned some 45 years. I first came under his influ- ence as a high school student when he took me on a personally guided tour of his “tarantula hill” south of Fayetteville. He had at one time the burrows of as many as 80 Dugesiella hentzi marked in this locale, and he recorded their habits and life history data as the colony waxed and waned. About 25 years later he obligingly offered to share his laboratory with me and for nearly three years provided a daily fare of wit and wisdom. In the 1930’s he built a house on a spacious lot on the outskirts of Fayetteville upon which he collected and planted many of the lesser common species of Ozark trees and plants. He also transplanted some of the native tarantulas there for easily accessible observation. To avoid the disturbance of wandering dogs and children he surrounded the entire property with a four-foot high stone wall, one of the iron gate entrances of which was wrought in the form of an orb web with its spider in place. He outlived some of the trees he planted and all of his tarantulas, but the garden still retains its character and beauty. Such a long career as his may be essential to study the life history of such long-lived creatures. His is probably the only complete life history study of Dugesiella hentzi one which lasted 24 years from eclosion to death in the specimens he cultured. The considerable arachnological library that Professor Baerg had accumulated through his more than 60 years of work, containing many now-rare reprints, has been donated to the University of Arkansas Library where they are now bound and have become a part of that library’s rare book collection. William B. Peck LIST OF ARACHNOLOGICAL PUBLICATIONS - W. J. BAERG 1922. Regarding the habits of tarantulas and the effects of their poison. Sci. Monthly, 14(5):482-489. 1923. The black widow: Its life history and the effects of the poison. Sci. Monthly, 18(6) :5 35-547. 1923. The effects of the bite of Latrodectus mactans (Fabr.).J. Parasit., 9:161-169. 1925. The effect of the venom of some supposedly poisonous arthropods of the Canal Zone. Ann. Entomol. Soc. Amer., 18:471-478. 1926. Regeneration of appendages in the tarantula Eurypelma calif ornica Ausserer. Ann. Entomol. Soc. Amer., 19(4) :5 1 2-5 1 3. 1928. Some studies of a trapdoor spider (Araneae: Aviculariidae). Entomol. News, 39(1): 1-4. 1928. The life cycle and mating habits of the male tarantula. Quart. Rev. Biol. 3(1): 109-1 16. 1929. Some poisonous arthropods of North and Central America. Trans. IV Internatl. Congr. En- tomol., 2:418-438. 1929. Cocoon making by the tarantula. Ann. Entomol. Soc. Amer., 22(2): 161-167. 1934. Some poisonous arthropods of southwestern Mexico. Ann. Entomol. Soc. Amer., 27:527-532. 1936. The black widow. Univ. Arkansas Agri. Expt. Sta. Bui., 325. 34 p. 1938. Tarantula studies. J. New York Entomol. Soc., 46:3143. 1945. The black widow and the tarantula. Trans. Connecticut Acad. Arts Sci., 36:99-113. 1948. The black widow. Pests (for April): 16-17. 1954. The brown widow and the black widow in Jamaica. Ann. Entomol. Soc. Amer., 47(l):52-60. 1954. Regarding the biology of the common Jamaican scorpion. Ann. Entomol. Soc. Amer., 47(2):272-276. 1955. The black widow’s little sister. Today’s Health, 33:42-43. 1958. The tarantula. Univ. Kansas Press, Lawrence, 88 p. 1959. The black widow and five other venomous spiders in the United States. Univ. Arkansas Agri. Expt. Sta. Bui., 608. 42 p. 1961. Scorpions: Biology and effect of their venom. Univ. Arkansas Agri. Expt. Sta. Bui., 649. 34 p. THE AMERICAN ARACHNOLOGICAL SOCIETY President: Herbert W. Levi (1979-1981) Museum of Comparative Zoology Harvard University Cambridge, Massachusetts 02138 Membership Secretary: Norman I. Platnick (appointed) American Museum of Natural History Central Park West at 79th Street New York, New York 10024 Secretary: William A. Shear (1980-1982) Department of Biology Hampden-Sydney College Hampden-Sydney, Virginia 23943 President-Elect: Jonathan Reiskind (1979-1981) Department of Zoology University of Florida Gainesville, Florida 32601 Treasurer: Norman V. Horner (1979-1981) Department of Biology Midwestern State University Wichita Falls, Texas 76308 Directors: Charles D. Dondale (1979-1981) Susan E. Riechert (1979-1981) Vincent D. Roth (1980-1982) The American Arachnological Society was founded in August, 1972, to promote the study of the Arachnida, to achieve closer cooperation between amateur and professional arachnologists, and to publish The Journal of Arachnology. Membership in the Society is open to all persons interested in the Arachnida. Annual dues are $20.00 for regular members, $12.00 for student members. Correspondence con- cerning membership in the Society must be addressed to the Membership Secretary. Members of the Society receive a subscription to The Journal of Arachnology. In addi- tion, members receive the bi-annual newsletter of the Society, American Arachnology. American Arachnology , edited by the Secretary, contains arachnological news and comments, requests for specimens and hard-to-Find literature, information about arach- nology courses and professional meetings, abstracts of the papers presented at the Society’s meetings, address changes and new listings of subscribers, and many other items intended to keep arachnologists informed about recent events and developments in arach- nology. Contributions for American Arachnology must be sent directly to the Secretary of the Society. The Eastern and Western sections of the Society hold regional meetings annually, and every three years the sections meet jointly at an International meeting. Information about meetings is published in American Arachnology , and details on attending the meetings are mailed by the host(s) of each particular meeting upon request from interested persons. The next International meeting will be held during 5-7 August 1981, hosted by Dr. Susan E. Riechert, Department of Zoology, University of Tennessee, Knoxville, TN 37916, USA. CONTENTS THE JOURNAL OF ARACHNOLOGY VOLUME 9 WINTER 1981 NUMBER 1 Feature Articles Revision of the genus Trachyrhinus Weed (Opiliones, Phalangioidea), James C. Cokendolpher The order Schizomida (Arachnida) in the New World. IV . goodnightorum and briggsi groups and unplaced species (Schizomidae, Schizomus),J. Mark Rowland and James R. Reddell Cavernicolous species of Larca, Aracheolarca and Pseudogarypus with notes on the genera (Pseudoscorpionida, Garypidae and Pseudogarypidae), William B. Muchmore Revision del genero Phiale C. L. Koch, 1846 (Araneae, Salticidae) III. Las especies polimorficas del grupo mimica, Maria E. Galiano Nest-mediated sexual discrimination by a jumping spider ( Phidippus fohnsoni), Robert R. Jackson Scorpions of the gems Paruroc tonus from New Mexico and Texas (Scorpiones, Vaejovidae), W. David Sissom and Oscar F. Francke Research Notes Report of necrophagy in the Black Widow Spider, Latrodectus Hesperus (Araneae, Theridiidae), Kenneth G. Ross 109 Redescription of Chthonius virginicus Chamberlin (Pseudoscorpionida, Chthoniidae), William B. Muchmore 110 Status of Leiobunum serratipalpe Roewer (Opiliones, Leiobunidae), James C. Cokendolpher 112 Book Review Les Araignees: Generalites— Araignees de France et des Pays Limitrophes, par Michel Hubert, Charles D. Dondale 114 : V Obituary W. J. Baerg, 1885-1980, William B. Peck Cover illustration, Schizomus pentapeltis (Cook), by W. D. Sissom Printed by The Texas Tech University Press, Lubbock, Texas Posted on April 1981 tCC The Journal of ARACHNOLOGY OFFICIAL ORGAN OF THE AMERICAN ARACHNOLOGICAL SOCIETY VOLUME 9 SPRING 1981 NUMBER 2 THE JOURNAL OF ARACHNOLOGY EDITOR’. Oscar F. Francke, Texas Tech University. ASSOCIATE EDITOR’. B. J. Kaston, San Diego State University. EDITORIAL ASSISTANT: James C. Cokendolpher, Texas Tech University. EDITORIAL BOARD: Charles D. Dondale, Agriculture Canada. William G. Eberhard, Universidad de Costa Rica. Maria E. Galiano, Museo Argentino de Ciencias Naturales. Willis J. Gertsch, American Museum of Natural History. Neil F. Hadley, Arizona State University. B. J. Kaston, San Diego State University. Herbert W. Levi, Harvard University. Emilio A. Maury, Museo Argentino de Ciencias Naturales. William B. Muchmore, University of Rochester. Martin H. Muma, Western New Mexico University. William B. Peck, Central Missouri State University. Norman I. Platnick, American Museum of Natural History. Susan E. Riechert, University of Tennessee. Michael E. Robinson, Smithsonian Tropical Research Inst. Jerome S. Rovner, Ohio University. William A. Shear, Hampden-Sydney College. Carlos E. Valerio, Universidad de Costa Rica. Stanley C. Williams, San Francisco State University. THE JOURNAL OF ARACHNOLOGY is published in Winter, Spring, and Fall by The American Arachnological Society in cooperation with The Graduate School, Texas Tech University. Individual subscriptions, which include membership in the Society, are $20.00 for regular members, $12.00 for student members. Institutional subscriptions to The Journal are $25.00. Correspondence concerning subscription and membership should be addres- sed to the Membership Secretary (see back inside cover). Back issues of The Journal are available from Dr. William B. Peck, Department of Biology, Central Missouri State Univer- sity, Warrensburg, Missouri 64093, U.S.A., at $5.00 for each number. Remittances should be made payable to The American Arachnological Society. Change of address notices must be sent to both the Secretary and the Membership Secretary. Manuscripts for THE JOURNAL OF ARACHNOLOGY are acceptable only from cur- rent members of the Society, and there are no page charges. Manuscripts must be type- written double or triple spaced on 8.5 in. by 11 in. bond paper with ample margins, and may be written in the following languages: English, French, Portuguese, and Spanish. Contributions dealing exclusively with any of the orders of Arachnida, excluding Acari, will be considered for publication. Papers of a comparative nature dealing with chelicer- ates in general, and directly relevant to the Arachnida are also acceptable. Detailed instructions for the preparation of manuscripts appear in the Fall issue of each year, and can also be obtained from the Editor and the Associate Editor. Manuscripts that do not follow those instructions will be returned to the author(s) without the benefit of review. Manuscripts and all related correspondence must be sent to Dr. B. J. Kaston, 5484 Hewlett Drive, San Diego, California 92115, U.S.A. Quintero, D. 1981. The amblypygid genus Phrynus in the Americas (Amblypygi, Phrynidae). J. Arachnol., 9:117-166. THE AMBLYPYGID GENUS PHRYNUS IN THE AMERICAS (AMBLYPYGI, PHRYNIDAE) Diomedes Quintero Jr. Director, Museo de Invertebrados Escuela de Biologia, Universidad de Panama Panama, Republica de Panama ABSTRACT Collections of Phrynus from all available localities from the Americas have been examined. Collec- tions from other parts of the world tropics have also been examined in the search for a better characterization of the three families of amblypygids which lack pulvilli in the tarsi of their ambula- tory legs. The gensus Phrynus has been found to occur only in the Americas and a redefinition of the Phrynidae is given. Of the 24 species of Phyrnus described for the Americas, 12 species are considered as valid and redescribed. The three species described by Franganillo, Phrynus pinarensis, P. viridescens and P. rangelensis are considered species Incertae Sedis. Three new species are described; one, P. damonidaensis , presents a character unique within the family Phrynidae: three tibial segments in leg IV instead of four. Five new synonymies are recognized. Natural history data are presented for some of the species of Phrynus. INTRODUCTION Phrynids are medium- to large-sized amblypygids (maximum body length about 45 mm) which occur in the tropical and semitropical areas of the Americas. Their habits are little known. Some are strictly cavernicolous (troglobites), others epigean, and a third group could colonize either above- or underground habitats. Of the three genera of phrynids, Paraphrynus Moreno appears to have a predominance of cave dwellers (troglo- philes) and the only troglobite species in the Phrynidae (Mullinex 1975, 1979). Hetero- phrynus Pocock and Phrynus Lamarck have both epigean and cave dwellers, and Acantho- phrynus Kraepelin, with a single known species, A. coronatus (Butler), appears to be strictly epigean of habits. Sexual dimorphism manifested as a positive allometric growth of the pedipalp length in males, not seen in females is known only from some species of Heterophrynus and Paraphrynus (Quintero 1979). The genus Phrynus is of interest because of its wide distribution in the Americas and the Caribbean islands, and the numerous poorly characterized species that have been named. The primary objectives of this study were a revision of the genus Phrynus and a search for new taxonomic characters to differentiate species; also to compile the known information and add new data pertaining to their biology. 118 THE JOURNAL OF ARACHNOLOGY Up to the present, the subfamily Phryninae, here considered a family as did Pocock (1902b), had been defined soley by the presence of four tibial segments on leg IV. All amblypygids without pulvilli on their tarsi and with four tibial segments on leg IV were considered phrynids. One of the most interesting findings of the present work has been the challenge of this narrow definition of the Phrynidae, finding a new species, Phrynus damonidaensis , with only three tibial segments on leg IV, but clearly ascribable by other characters to the Phrynidae and the genus Phrynus and not to the Damonidae. Phrynus santarensis (Pocock) may represent a second case not yet confirmed with this unique segmentation (see discussion under P. santarensis). These findings brought a redefinition of the Phrynidae comparable to the redefinition of the Damonidae (see Quintero 1976). The Phrynus - Tarantula name polemic is one of the longest standing polemics in zoological nomenclature. It started with the opinion that Linnaeus used for the descrip- tion of his only amblypygid species, Phalangium reniforme L., 1758, the drawing of a specimen from Antigua, West Indies, made by Browne (1789). In fact, it was later demon- strated that Linnaeus had one specimen from the Orient at the Zoological Museum of the Royal University at Uppsala upon which he based his description (Lonnberg 1897). It was renamed Phrynichus reniformis by Karsch (1879) upon removal from the opilionid genus Phalangium (type-species of Phrynichus by original designation). In the same work, Phalangium reniforme L. became, both by subsequent designation, type-species of Taran- tula Fabricius and Phrynus Lamarck. The amblypygid genus Phrynus Lamarck, 1801 was the second established genus with- in the order Amblypygi, after Tarantula Fabricius, 1793, and was the genus used by different authors to accomodate all new taxa for nearly one century, from 1801 to 1893 (e.g., Butler 1873). In 1893 Pocock revived the name Tarantula to replace Phrynus for American species of amblypygids, ignoring Karsch’s work (1879). For Pocock (1893, 1894), until 1902 Tarantula was an unresolved blend of Phrynus and Paraphrynus . When he resolved this blend in 1902, separating Phrynus from the then-named Hemiphrynus, Pocock used the name Phrynus. In a recent petition to the International Commission on Zoological Nomenclature, I have resumed the polemic Phrynus - Tarantula and asked the Commission to preserve the generic name Phrynus Lamarck, 1801 and suppress the generic name Tarantula Fabricius, 1793 for the purposes of the Law of Priority, but not for those of the Law of Homonymy, and place it on the Official Index of Rejected and Invalid Generic Names in Zoology (Quintero, in press). Admetus C. L. Koch. 1850, and Neophrynus Kraepelin, 1895 then fall as junior objective synonyms of Phyrnus. The conclusions attained best serve stability of nomenclature for the following rea- sons: a. Tarantula is hopelessly compromised, having been used with a variety of meaning in Hexapoda and even for a fish, and being associated in the vernacular with a genus far removed from the Amblypygi. b. Phalangium reniforme L., 1758, is now generally used for the eastern species to which Lonnberg assigned it and for which the generic name Phrynichus is generally used (for references see Delle Cave and Simonetta 1975). TAXONOMIC METHODS For the selection of the diagnostic characters I prepared first a list and checked the taxomonic characters used by Pocock and those pointed out by Weygoldt (1970). In QUINTERO-THE AMBLYPYGID GENU S.PHRYNUS 119 addition, I later added the cheliceral teeth as suggested by Mullinex for Paraphrynus (1975), a series of measurements, and calculated positional ratios for some important trichobothria. The present study is based on the features of preserved specimens, hence species criteria are strictly morphological. The pedipalp spination is very important and full descriptions are given for each species. Names of the different articles follow those used by Weygoldt (1970). For the first time in relation to Phrynus the cheliceral denti- tion, genitalia of males and females, and trichobothria on leg IV have been illustrated. Drawings and measurements.— A few drawings are presented to show the general appearance and structure of the genus (Figs. 1-6). Specimens were measured and illustrated by the use of a standard ocular grid in a binocular dissecting microscope. The magnification varied and thus the limits of accuracy. For larger dimensions the measure- ments are accurate to ± 0.1 mm; for the trichobothria, features are accurate to ± 0.01 mm. Drawings and measurements were made with the segment in as nearly horizontal a plane as possible. Illustrations were made using magnifications of 60X to 120X for the pedipalps and trichobothria and 250 X for genitalia and cheliceral teeth. Measurements were made as indicated in Figures 14-17. Figs. 1-5 -Phrynus tessellatus (Pocock), from St. Vincent: 1, dorsal view of carapace and left chelicera; 4, inner view of pedipalp tarsus. Phrynus gervaisii (Pocock), from Madden Forest Preserve, Canal Zone: 2, ventral view of left pedipalp and chelicera; 3, dorsal view of left pedipalp; 5, part of distitibia and whole tarsus of left leg IV. Abbreviations: UCS, upper celiceral segment; ST, setiferous tubercle; BCS, basal cheliceral segment; FP, frontal process; FA, frontal area; S, sulcus of carapace; TA, tarsus; BT, basitarsus; T, tibia; F, femur; TR, trochanter; G, gnathocoxa; CO, cleaning organ; SP, spine on dorso-inner lateral surface of pedip^ip tarsus; D, distitibia. 120 THE JOURNAL OF ARACHNOLOGY Pedipalp spination.— Previous work has used almost exclusively the pedipalp spination to separate species. Although the pedipalps are still considered as suppliers of the main diagnostic characters, additional taxonomic characters have been used to sort species and the description of the pedipalp spination has been made more accurate by numbering the spines and illustrating them, following Mullinex (1975). The most reliable spines on the pedipalp appear to be dorsal on the tibia. The following abbreviations were used when describing the spination of the pedipalps: Fd - femur, dorsal surface; Fv - femur, ventral surface; Td - tibia, dorsal surface; Tv - tibia, ventral surface; Bd - basitarsus, dorsal surface; Bv - basitarsus, ventral surface. Following each abbreviation there is a number which indicates the position of the spine. The spines which I found to be more constant were numbered progressively from proximal to distal end of each pedipalp segment. For example, Td-6 is the 6th spine from the proximal end found on the dorsal surface of the tibia of the pedipalp. Equal numbers do not necessarily mean homologous spines on the same segment. The actual fifth spine on the femur in one individual of a species may not be homologous with the fifth spine on the same segment in another species. Some shorter spines numbered may cause confusion and for this reason it is preferable to refer frequently to illustrations when reading the descriptions. Cheliceral dentition.— It gives valuable information to differentiate species, particularly the teeth on the external margin of the basal cheliceral segment. The right chelicera was dissected and the inner rows of hairs on the basal cheliceral segment that block the view of the external teeth were removed. No characters of diagnostic value for Phrynus species were found on the medial surface of the basal cheliceral segment. In the family Phrynidae, on this surface Acanthophrynus coronatus has a stridulatory apparatus (Shear 1970) and Paraphrynus astes Mullinex only clavate setae, probably also stridulatory in function (Mullinex 1975). Specimens of nine species of Heterophrynus were examined and found to lack either a stridulatory appara- tus or clavate setae. Genitalia.— Although it is possible to distinguish species without using genitalic char- acters, the external female genitalia appear as a source of potentially useful additional characters to distinguish species and should be used in future revisions. I preferred not to introduce newly coined words to describe it and instead have given dorsal illustrations which could serve to help to corroborate identifications. Compared with the male geni- talia, they are simpler and easier to study. The sclerites are hard and not likely to change in shape during preservation. The male genitalia are less useful for specific diagnosis, being much more complex and individually subject to frequent variations during preservation, because of their generally soft, muscular nature. Their pale anatomy is difficult to inter- pret. The dorsal and ventral illustrations of the male genitalia are presented. For the illustrations of the genitalia, it was necessary to dissect the genital operculum from specimens. According to Weygoldt (1972, 1975), the pair of limb buds of the eight embryonic segment (second opisthosomal) develops posteriorly the first pair of book lungs and. largely from the upper part of those limb buds, the genital operculum and the genital appendages (gonopods of males and females). Thus Weygoldt (1972, 1975) homologized the erectile bodies of the male and female genitalia with paired abdominal appendages (extremities), homologous to the telopodites and endopodites of the opisthosomal extremities of eurypterids, while the book lungs were considered to have evolved from praepodites. I will refer to the female genitalia, accepting Weygoldt’s interpretations, as female gonopods, and will coin a new name for the male genitalia, frequently referred to QUINTERO-THE AMBLYPYGID GENU S PHR YNUS 121 incorrectly as penis, notwithstanding their noncopulatory, spermatophore-producing function. The male organs of amblypygids will by named “opisthogeminate organs,” from the Greek opisthen , meaning backwards, and the Latin geminatus, meaning paired, refer- ring to the biconic, backwards protusible organs located posteriorly on the dorsal side of the genital operculum of male amblypygids. Color.— Color descriptions are based upon alcohol-preserved specimens. Amblypygids retain color well in alcohol, as I have found in comparing old museum specimens with recently preserved ones. Coloration patterns of the abdominal tergites proved useful as diagnostic characters, as did the banding on the femora of the ambulatory legs. Trichobothria.-The trichobothria present on leg IV have been an additional character used for the description of species. Lamentably, examination is difficult and awkward because of their position on the distitibia of leg IV. Trichobothria also frequently present anomalies in their position and arrangement, which makes interpretations more difficult. The information content of future taxonomic work on Phrynus can be increased by inclusion of additional data on the trichobothria. At present, an understanding of variabil- ity is needed. Species like Phrynus whitei Gervais present frequent and significant varia- tions in their trichobothria numbers and ratios, while species like Phrynus gervaisii (Po- cock), for example, present numbers and ratios rather constant when specimens from Panama are compared with those from several different localities in South America. The following abbreviations have been used to describe the trichobothria present in each segment of leg IV (Figs. 160-174). Trichobothria present pt - proximotibial none bt - basitibial bf - basofrontal be - basocaudal sbf - subbasofrontal stf - subterminofrontal sbe - subbasocaudal sci-x • series caudal and trichobothria present sfi-x - series frontal and trichobothria present The group of three trichobothria placed below the sc and sf rows is the terminal group: tm, terminal medial; tf, terminal frontal; tc, terminal caudal. They are constant in all members of the order Amblypygi and do not show variation. In Phrynus damonidaenis and P. santarensis (Pocock), the second tibial segment (pre- basitibia) of leg IV is not present but, as this segment does not have trichobothria, its pattern appears normal. During development, trichobothria numbers are reduced (Weygoldt 1970). The tricho- bothria bt, bf and be are double in immatures (with distal and caudal trichobothria each) but in adults, which are the ones represented in the drawings of the present work, there is only one of each of those trichobothria. The number of trichobothria from the sc and sf rows is also reduced and those trichobothria still present have been numbered from the proximal to the distal end of each row. For example, for Phrynus levii sc 1,2, 6-11 means that hairs 3 to 5 are missing but hairs 1, 2, and 6 to 1 1 are present on the caudal series. Tibia IV segments 1 . Proximal tibia 2. Pre-basitibia 3. Basitibia 4. Distitibia (old names: metatarsus, basitarsus) 122 THE JOURNAL OF ARACHNOLOGY In order to better locate trichobothria on each segment and to be able to compare quantitatively their positions, I have calculated ratios for the following: pt, bt, bf, sbc, and sci. Each ratio is calculated by dividing the distance of the trichobothria to the proximal end of the segment by the total length of the segment. Higher ratios indicate that the trichobothria is more distally positioned. Segmentation of Antenniform Leg.— The number of segments present on leg 1 (anten- niform leg), although subject to frequent abnormalities in segmentation, is rather con- stant for each species. I have found four groups of species based on the number of segments of leg 1 : 1. With 25 tibial segments, patella not included. Only P. parvulus Pocock belongs to this group. It has 57 tarsal segments and a total of 82 segments. 2. With 27 tibial segments. Includes three species: P.levii, n.sp., P. damonidaensis, n.sp., and P. margine macula tus C.L. Koch. They all have 59 tarsal segments and a total of 86 segments. 3. With 29 tibial segments. Includes three species: P. operculatus Pocock, P. asperatipes Wood, and P. whitei Gervais. The numbers of tarsal segments varies from 60 to 63, and the total number of segments from 89 to 92. It is interesting to point out that Acanthophrynus coronatus (Butler) has a similar number of segments (Quintero 1980). 4. With 31 tibial segments. All other known species belong to this group. The number of tarsal segments varies from 66 to 68 and the total number of segments from 97 to 99. Types depositories.— Type of two new species, P. armasi and P. damonidaensis , are being deposited in the Academia de Ciencias de Cuba (ACC). The lectotype of P. goesii Thorell is at the Naturhistoriska Ricksmuseum of Stockholm. The neotype of P. aspera- tipes Wood is in the Museum of Comparative Zoology (MCZ), Harvard University, and all other type specimens are deposited in the British Museum (Natural History) (BMNH). THE DEFINITION OF THE PHRYNIDAE The family name Phrynidae, as first used by Wood (1863b) to include the single genus Phyrynus , ranked at the now known level of order. Gervais had previously (1844) created the order Phryneides, with the single genus Phrynus. The name Phrynides was used later for the order rank, and credit was given to Gervais. The name Phrynidae has been cited incorrectly as Prynoidae by L. Koch and as Phrynoidae by Thorell (1889). The second amblypygi family name was Karsch’s Tarantuliden (1879), later used as Tarantulidae by Simon (1892) and others. Simon (1892) was the first to subdivide the family Taran- tulidae in three subfamilies: Charontinae, Phryniscinae (sic) and Tarantulinae. Kraepelin (1895) renamed Tarantulinae the Phrynichinae of Simon (for oriental species), and Neo- phryninae the Tarantulinae of Simon (for American species). Kraepelin, in 1899, returned to the use of Simon’s names, and included in Tarantulinae Simon the genera A dmetus C. L. Koch 1841, Acanthophrynus new name, and Tarantula Fabricius 1793. Pocock (1902a,b) returned to the older name Phrynidae to elevate in rank the subfamily Taran- tulinae, and subdivided it into two subfamilies: Phryninae and Heterophryninae. Pocock (1902b) also elevated to family rank the other subfamilies of Simon: Tarantulidae and Charontidae. His work was ignored and posterior usage was of a single family, Taran- tulidae, for the whole order Amblypygi. The family name Phrynidae remained unused QUINTERO-THE AMBLYPYGID GENU S PHR YNUS 123 until 1975, when Mullinex mentioned it in her work also without designating a type genus. According to the ICZN, art. 23c, the name Phrynidae must be attributed to Wood and not to Pocock, who indicated erroneously he was the author of the name (1902b). Family Phrynidae Wood Phrynidae Wood 1863b: 375; Pocock 1902b: 157-165. Tarantulinae Simon 1892:50 (in part); Pocock 1894:273; Kraepelin 1899:240. Neophryninae Kraepelin 1895:8. Admetinae Pocock 1897:358. Diagnosis.— Amblypygids without pulvilli on the tarsi of ambulatory legs (see Quintero 1975) and with the tibia of leg IV divided into four or three segments; with the proximal cusp of the inner double pointed proximal tooth (Fig. 7, arrow) on the median edge of the basal cheliceral segment being shorter than the distal cusp on the same tooth; the basocaudal row of trichobothria on the distitibia of leg IV is not present. They have basocaudal, basofrontal, subbasocaudal, subbasofrontal and subterminal trichobothria in addition to the caudal and frontal series. The angle of the articulation on the pedipalp Figs. 6-1 3.-6, sternites of Phrynus gervaisii (Pocock) from Panama; 7, arrow indicates the smaller proximal cusp on the inner double-pointed tooth, typical of all members of the Phrynidae; 8, Acan- thophrynus coronatus (Butler) of Colima, Mexico, basitarsus and tarsus, inner-lateral view; 9, articu- lation of pedipalp trochanter-femur, typical of Phrynidae; 10, atriculation of pedipalp trochanter- femur of Damonidae; 11, ventral of pedipalp trochanter with sub cylinder ical sclerotized apophysis projecting posteriorly; 12, anterior edge of carapace of A cant ho phrynus coronatus (Butler), with long spiniform processes; 13, dorsal of pedipalp tibia of Paraphrynus laevifrons (Pocock) from Panama. 124 THE JOURNAL OF ARACHNOLOGY between trochanter and femur is always median and not lateral to the line of dorsal spines on the femur in Phrynus and Paraphrynus (Fig. 9) but dorsally displaced in Hetero- phrynus and Acanthophrynus. Compare their type of articulation with the type present in the Damonidae (Fig. 10), which is similar to the one present in the Phrynichidae, where the articulation trochanter-femur is laterodorsal. I hereby designate the genus Phrynus as the type genus of the Phrynidae. It is the oldest genus within the family and a member of the subfamily Phryninae which lacks the subcylindrical sclerotized apophysis on the ventral surface of the pedipalp trochanter. This apophysis is present in the other subfamily, Heterophryninae (the only genus includ- ed, Heterophrynus). KEY TO GENERA OF PHRYNIDAE 1. Ventral surface of the pedipalp trochanter with subcylindrical sclerotized apophysis projecting posteriorly (Fig. 11) Heterophrynus Ventral surface of the pedipalp trochanter without such apophysis (Fig. 2) ... 2 2. Anterior edge of carapace with long spiniform processes (Fig. 12); basitarsus of pedi- palp armed with a single long dorsal and ventral spine (Fig. 8) . . . . Acanthophrynus Anterior edge of carapace armed with short denticuliform tubercules or almost smooth (Fig. 1); basitarsus of pedipalp armed, at least dorsally, with more than one long spine (Fig. 3) 3 3. Dorsal margin of pedipalp tibia with two spines between the longest spines (Fig. 13) Paraphrynus Dorsal margin of pedipalp tibia with one spine (Td-4) between the two longest spines (Fig. 3) Phrynus Genus Phrynus Lamarck Tarantula Fabricius 1793:432 (misidentification of Phalangium reniforme L.), Karsch 1879:197 (misi- dentification of Phrynus pumilio C. L. Koch, known to be a species of Heterophrynus ), Pocock 1893:527 (in part), 1894:275 (in part), Kraepelin 1899:241 (in part). Type-species Phalangium reniforme L. 1758, syntype lost from Mus. Lud. Ulr. Lonnberg, 1897, virtually designated the syntype as lectotype of P. reniforme L. and identified it as an East Indian species of Phrynichus. Phrynus Lamarck 1801:175 (in part), C. L. Koch 1841:13 (in part), Gervais 1842:19 (in part), Butler 1873:118 (in part), Pocock 1902a:50, 1902b:161. Phalangium palmatum Herbst 1797, type- species, subsequent designation by Pocock 1902a: 50 and 1902b:161. Previous type-species desig- nations very obscure (see Quintero, in press). After lengthy history of misidentifications, having its type lost, presumedly destroyed, I have reassessed Herbst descriptions and illustrations and found it to be a synonym of Phrynus operculatus Pocock, 1902a. See farther discussions under “On the Identity of P. palmatum Herbst”. Admetus C.L. Koch 1850:81 (in part), Pocock 1897:358. Type-species Phalangium palmatum Herbst, subsequent designation by Simon 1892:51. Neophrynus Kraepelin 1895:23 (in part). Type-species by original designation Phalangium palmatum Herbst (misidentification). N. palmatus (Herbst) was a polytypic species formed by lumping seven distinct species. Diagnosis.— Anterior edge of carapace at most weakly denticulated. Pedipalp tibia armed dorsally and ventrally only with marginal spines which tend to cluster distally only in the adult species of Phrynus whitei; between the two longest spines (Td-3 and Td-5) QUINTERO-THE AMBLYPYGID GENUS PHRYNUS 125 there is one shorter spine (Td-4). This single spine distinguishes Phrynus from the closely related genus Paraphrynus which presents two spines between the two longest spines. Basitarsus armed dorsally with two to three long spines and ventrally with one to three long spines. The maximum body length measured was 35 mm. The only species of Acanthophrynus, A. coronatus, reaches a maximum length of 45 mm and has a different basitarsus and frontal edge spination on the carapace. matum Herbst by subsequent designation. Figs. 14-1 7. -Diagram of measurements taken: 14, a = distance of median ocular tubercle from anterior edge; b = distance between lateral eyes; c = length of carapace; d = width of carapace; e = distance of lateral eyes from anterior edge; f = distance of lateral eyes from lateral edge. 15, g = length of left pedipalp tibia; h = width of pedipalp tibia. 16, i = length of pedipalp tarsus; j = length of pedipalp basitarsus; k = width of pedipalp basitarsus. 17,1 = length of left pedipalp femur. 126 THE JOURNAL OF ARACHNOLOGY Description-Carapace: rather uniform in shape, broader than long, moderately high with a distinct sulcus variable in shape, mainly transversely oriented (Fig. 1). Body covered with sharp pointed setae and spines except in Phrynus parvulus Pocock which is covered with clubiform setae and blunt spines. Chelicera: consisting of two segments, a basal and a clawlike projection, both articulated (Fig. 1). The clawlike projection or upper cheliceral segment presents a row of four to five teeth of variable length and shapes. The basal segment presents two constant rows of teeth on its anteroventral surface (Figs. 108- 122). The inner or medial row has always three teeth, the proximal one is the inner double pointed tooth. The distal cup of this tooth is always larger than the proximal one. The external edge of the basal cheliceral segment may have from one to four teeth. Distal on the anterodorsal edge of this segment, there is in some species a well-developed setiferous tubercle. Frontal process: sclerotized projection of variable shape located above the entrance to the buccal cavity that could be large and project from under the anterior edge and become visible from above. Eyes: a median ocular tubercle well developed in all species, the lateral eye clusters with frequent abnormalities. Variable distance of median ocular tubercle and lateral eye clusters to anterior and lateral edges of carapace. Sternum: prostemum robust, projecting between the base of the gnathocoxae. Meso and pentasternum reduced. Very reduced articulation scleriteson the sternal plate at the points of articulation with the coxae of ambulatory legs (Fig. 6). Genitalia: female and male gonopods originating from the inner wall of the genital oper- culum (Figs. 123-159). In the female there are two protrusible organs, the female gono- pods, each bearing a clawlike sclerite pointing backwards and inwards. Lateral to each sclerite and between the gonopods there are sclerotization areas variable between the species. The male gonopods, the opisthogeminate organs, are pale, muscular and complex, ending each half in a cone posteriorily. Pedipalp: dorsal surface of pedipalp tibia with seven to nine spines. Spine Td-4 always shorter than Td-3 and Td-5, which are the two longest spines on the pedipalp. Tarsus with a cleaning organ consisting of two rows of short bristles (Fig. 4). The dorsomedial row of minute bristles is absent from a single species, Phrynus asperatipes. Besides the cleaning organ there is present in some species a small inconspicuous spine situated proximally on the dorso-inner lateral surface of the pedipalp tarsus (Fig. 4). The tarsus and posttarsus are fused except in P. asperatipes where a suture separates these two areas (Fig. 20). Legs: the tibia of leg IV presents normally from four to three segments. Rarely an abnormal segmentation of two segments is found. The second tarsomere may present a membranous transverse line distally, extending the width of the segment (Fig. 5). It could be absent in some species. Trichobothria: the ratios of pt, bt, bf, sbc and sci present interesting variations, as do the numbers of the sc and sf rows. KEY TO SPECIES OF PHRYNUS 1. A distinct dorsoventral suture line between pedipalp tarsus and post-tarsus (Fig. 20), cleaning organ without dorsomedial row of minute bristles (Fig. 20) . . asperatipes Pedipalp tarsus and post-tarsus fused, no suture line separating these two areas (Fig. 27, 32, 39); cleaning organ with dorsomedial row of minute bristles (Fig. 4) . . 2 2. Proximal end of dorso-inner lateral surface of pedipalp tarsus with small inconspic- uous spine (Figs. 4, 27, 32, 39, 44, 51) 3 Proximal end of dorso-inner lateral surface of pedipalp tarsus without inconspicuous spine (Figs. 56, 63, 68, 75, 80, 87) 7 QUINTERO-THE AMBLYPYGID GENUS PHRYNUS 127 3. One or three cheliceral teeth on external margin of basal cheliceral segment (Figs. 108, 109, 114, 115), no spine on anterior trochanter face or a short setiferous tubercle not larger than largest tubercle on anterolateral row (Figs. 24, 47) ... 4 Two teeth on external margin of basal cheliceral segment (Fig. 117); a well-developed spine or tubercle on the anterior face of the trochanter, always distinctly larger than the largest tubercle present on the anterolateral row (Fig. 35) pulchripes 4. One tooth on external margin of basal cheliceral segment (Fig. 118); Td-7 with a basal spine (Fig. 26) xirmasi Three teeth on external margin of basal cheliceral segment (Figs. 114-116); Td-7 without a basal spine (Figs. 38, 43, 50) 5 5. Td-6 with a basal spine (Fig. 43); Td-5 same size or shorter than Td-3; Td-4 longer or same size as Td-2 goesii Td-6 without a basal spine (Figs.. 38, 50); Td-5 shorter than Td-3; Td-4 shorter than Td-2 .6 6. Very well-developed frontal process (Fig. 1); femur of leg 1 is 2.3 to 2.7 times longer than medial prosomal length; sbc trichobothria higher than 0.53 (Fig. 170); Bv-1 longer than Bv-3 (Fig. 51) tessellatus Frontal process concealed; femur of leg 1 is 2.8 to 3.8 times longer than median prosomal length; sbc trichobothria lower than 0.53 (Fig. 172); Bv-1 shorter or same size as Bv-3 (Fig. 39) longipes 7. Td-4 shorter than Td-2 8 Td-4 longer than Td-2 13 8. Anterior face of pedipalp trochanter without a spine (Figs. 59, 60, 71); Bd-1 a very small, inconspicuous spine (Figs. 56, 63, 68) 9 A well-developed spine at the center of the anterior face of pedipalp trochanter (Figs. 72, 83, 84); Bd-1 a well-developed, distinct, spine at base of Bd-2 (Figs. 75, 80, 87) 11 9. Three tibial segments on leg four (Fig. 164) damonidaensis Four tibial segments on leg four (Figs. 161,1 66) 10 10. Two teeth on external margin of basal cheliceral segment (Fig. 120); dorsal of chelic- eral basal segment lighter than frontal area, rarely of same color, never darker; Fv-1 and Fv-2 with a common base, distinct from Fv-3 (Fig. 64) . . . marginemaculatus Three teeth on external margin of basal cheliceral segment (Fig. 121); dorsal of cheliceral basal segment darker than frontal area; separate base for each Fv-1, Fv-2 and F-3 (Fig. 69) levii 11. Five spines on the edges of the pedipalp trochanter (Fig. 72); Bv-1 a very well- developed spine (Fig. 75); Td-5 shorter than Td-3 (Fig. 74); medial face of pedipalp tibia, completely granular santarensis Four spines on the edges of the pedipalp trochanter (Figs. 83, 84); Bv-1 short and inconspicuous (Figs. 80, 87); Td-5 distinctly longer than Td-3 (Figs. 79, 86); medial face of pedipalp tibia with a central smooth band 12 12. Raised, distinctly delimited frontal area; tergites of abdomen pale yellow-brown scarcely a trace of pattern; eight dorsal spines on the pedipalp tibia (Fig. 79); frontal process visible from above although vertically positioned; a short spine betweenFv-5 and Fv-6(Fig. 81) barbadensis Poorly defined frontal area, its limits not distinctly set apart from rest of carapace; tergites of abdomen variegated yellow and dark brown; only seven dorsal spines on pedipalp tibia (Fig. 86); frontal process concealed; without spine between Fv-5 and Fv-6 (Fig. 88) gervaisii 128 THE JOURNAL OF ARACHNOLOGY 13. Td-5 shorter than Td-3 (Fig. 91); Td-2 and Td-6 about same size; two bright spots between lateral eyes; distal displacement of dorsal spines on pedipalp tibia (Fig. 91) whitei Td-5 and Td-3 about same size (Figs. 98, 104); Td-2 distinctly longer than Td-6; without spots between lateral eyes; normal distribution of dorsal spines on pedipalp tibia, without distal displacement (Figs. 98, 104) 14 14. Fv-3 shorter than Fv-6 (Fig. 100); with marked banding on femur of ambulatory legs; clubiform setae on carapace, pedipalps and legs; fourth abdominal sternite straight in males; males with small genital operculum parvulus Fv-3 longer than Fv-6 (Fig. 103); legs of a single color, no banding; sharp pointed setae on carapace, pedipalps and legs; fourth abdominal sternite bent in males; males with large genital operculum operculatus Phrynus asperatipes Wood Figs. 18-23,119, 159, 171; Map 1 Phrynus asperatipes Wood 1 863a: 111. Holotype (sex not determined in original description) from Baja California, probably Baja California Sur, Mexico, deposited in Smithsonian Museum (USNM), lost. Examined topotypic specimens, designated a female neotype. Wood 1863b:375-6; Butler 1873:118. Neophrynus whitei: Kraepelin 1895:28-30 (in part). Tarantula whitei: Kraepelin, 1899:242-3 (in part). Figs. 18-29. -Pedipalps: Phrynus asperatipes Wood, female neotype, 18-23 \ Phrynus armasi, new species, male holotype, 24-29. Figs. 18, 25 = femur, dorsal view; 21, 28 = femur, ventral view; 19, 26 = tibia, dorsal view; 22, 29 = tibia, ventral view; 20, 27 = basitarsus and tarsus, inner lateral view; 23, 24 = trochanter, anterodorsal view. QUINTERO-THE AMBLYPYGID GENU S PHR YNUS 129 Neotype. -Female from La Paz, Baja California Sur, Mexico, 3 February 1965 (V. Roth) deposited in MCZ. Diagnosis. -Phrynus asperatipes is a distinctive species easily recognized by its pedi- palp tarsus that lacks the dorsomedial row of minute bristles on the cleaning organ and bacause of its apparent suture between tarsus and post-tarsus. It is also recognized by its yellow-brown color, not seen in other species of Phrynus. Description Female Neotype.— Carapace, pedipalps and legs predominantly yellowish- brown. Yellow frontal area with black median ocular tubercle. Orange-brown sulcus besides which there are four dark shallow grooves. Dorsum of abdomen with median orange-brown band, the muscular impressions and dorsal border of tergites are of same color, the rest is yellow. Total length 18.0 mm. Carapace. Wide, nearly straight, anterior edge. Frontal process concealed. Distance of median ocular tubercle from anterior edge less than double length of tubercle (0.5 mm/0.3 mm). Carapace 7.9 mm long, 10.1 mm wide, 4.1 mm sulcus from anterior edge. Median ocular tubercle 0.5 mm wide. Lateral eyes 3.7 mm from each other, 1.3 mm from anterior edge, 1 .0 mm from lateral edge. Chelicerae. Dorsal surface of basal segment without distal tubercles. A single tooth on external margin of basal cheliceral segment (Fig. 119) and two ridges. The most proximal ridge appears to connect the external tooth with the inner double-pointed tooth. Genital operculum. 2.4 mm long, 4.8 mm wide. Female gonopods as in Fig. 159. Abdomen 12.4 mm long. Figs. 30-41. -Pedipalps of male holotypes: Phrynus pulchripes (Pocock), 30-35 ; Phrynus longipes (Pocock), 36-41. Figs. 30, 37 = femur, dorsal view; 33, 40 = femur, ventral view; 31, 38 = tibia, dorsal view; 34, 41 = tibia, ventral view; 32, 39 = basitarsus and tarsus, inner lateral view; 35, 36 = trochan- ter, anterodorsal view. 130 THE JOURNAL OF ARACHNOLOGY Trichobothria. As in Fig. 171; ratios: sbc 0.54 and bt 0.25. Pedipalps. Figs. 18-23. Trochanter with four spines. A setiferous tubercle at center of anterior surface shorter than larger tubercles on antero-external row. Femur Fd-3 same size as Fd-2. Fv-4 same size as Fv-7. Fv-1, Fv-2 and Fv-3 each on a separate base. Tibia, Td-6 longer than Td-1. Td-4 shorter than Td-2, longer than Td-6. Td-3 longer than Td-5. Tv-4 longer than Tv-6. Basitarsus Bd-1 very small, inconspicuous, one-fifth the length of Bd-3. Bv-1 and Bv-3, obsolete, Bv-2 well developed and longer than Bd-3. Pedipalp tarsus and post-tarsus not completely fused, an apparent suture visible dorsally and ventrally. Femur 6.2 mm long; tibia 6.4 mm long, 2.4 mm wide, basitarsus 3.3 mm long, tarsus 3.5 mm long. Legs. Second tarsomere of all tarsi without light transverse line on distal end. Anten- niform leg: 16.7 mm, femur; 30.3 mm, tibia; 34.0 mm long, tarsus. Leg II: 10.8 mm, femur; 15.6 mm, tibia. Leg III: 11.6 mm, femur; 17.0 mm, tibia. Leg IV: 10.0 mm, femur; 16.3 mm (6. 8/1. 2/3. 3/5.0), tibia; 2.4 mm ( 1 . 1/0.4/ 0. 1/0.8), tarsus. Natural History.— It has been collected from creeks, a palm oasis, under rocks on a hillside and in a sand dune area. Distribution. -Mexico: Baja California Sur. Phrynus pulchripes (Pocock) Figs. 30-35, 117, 123, 124, 126, 168; Map 2 Tarantula pulchripes Pocock 1894:283-4, pi. 7, Fig. 6, male, female. Male holotype from Colombia, in the BMNH, examined. Very fragmented and discolored. Neophrynus palmatus barbadensis: Kraepelin 1895:30-34 (in part). Tarantula palmata : Kraepelin 1899:242-244 (in part). Phrynus pulchripes: Mello-Leitao 1931:40, 43. Hemiphrynus corderoi • Mello-Leitao 1946:1-2, pi. 1, Figs. 1-2, Female holotype from Caracas, Vene- zuela, in the Rio de Janeiro Museum, examined. NEW SYNONYMY. Diagnosis.— Within the group of species that have a small, inconspicuous spine on the proximal end of dorso-inner lateral surface of pedipalp tarsus, Phrynus pulchripes is easily recognized by being a medium-size species with only two teeth on the external margin of the basal cheliceral segment and a well-developed spine on the anterior face of the pedipalp trochanter. The female genitalia presents two robust, dark-brown sclerites with a distinct shallow curve near their distal end. Description, Male holotype -Carapace and pedipalps dark brown. With distinct fla- vous marginal spots on the carapace. Legs distinctly banded: three yellow bands on the femora of the posterior three pairs of legs. A flavous ring around the muscular impres- sions of the tergites. Total length 16.8 mm. Carapace. With narrow, slightly emarginate anterior edge, with small, even size tuber- cles. Frontal process concealed from above. Distance of median ocular tubercle from anterior edge shorter than length of tubercle (0.4 mm/0.54 mm). The frontal region gently sloped downward and forward, steeper below the lateral eyes. Carapace 6 6 mm long, 11.1 mm wide, 4.0 mm sulcus from anterior edge. Median ocular tubercle 0.8 mm wide. Lateral eyes 3.6 mm from each other, 1.3 mm from anterior edge 1.5 mm from lateral edge. Chelicerae. Anterodorsal surface of basal cheliceral segment with the external distal tubercle slightly enlarged. Two teeth on external margin of basal cheliceral segment (Fig. 117). QUINTERO-THE AMBLYPYGID GENU S PHR YNUS 131 Genital Operculum. 3.9 mm long, 5.9 mm wide. Opisthogeminate organ as in Figs. 123 and 124. Abdomen 1 1.8 mm long. Trichobothria. As in Fig. 168. Ratios: sbc 0.56, bt 0.50. Pedipalps. Figs. 30-35. Trochanter with four spines, a well-developed spine at center of anterior surface, larger than largest tubercles on antero-external row. Femur, coarsely granular above, Fd-2 longer than Fd-3, Fd-4 longer than Fd-5 and Fd-6. Fv-1, Fv-2 and Fv-3 on a common base. Fv-6 longer than Fv-4. Tibia, Td-6 longer than Td-1 but shorter than Td-2. Td-4 shorter than Td-2 and longer than Td-6. Td-3 longer than Td-5. Tv-4 and Tv-6 are the same size, and shorter than Tv-1. Basitarsus. Bd-1 long, longer than half the Map 1.- Distributions of Phyrnus arrmsi, P. asperatipes, P. damonidaensis, P. levii, P. longipes, P. marginemaculatus, P. operculatus, P. parvulus, and P. whitei. 132 THE JOURNAL OF ARACHNOLOGY length of Bd-3. A short spine distal to Bd-3, shorter than Bd-1. Bv-3 and Bv-1 are the same size. Bv-1 shorter but thicker than Bd-1. Bv-2 and Bd-3 are the same size. Pedipalp tarsus and post-tarsus completely fused. Tarsus with proximal end of dorso-inner lateral surface with small, inconspicuous spine. Femur 7.0 mm long; tibia 7.9 mm long, 2.6 mm wide; basitarus 3.7 mm long, 2.4 mm wide; tarsus 3.7 mm long. Legs. Second tarsomere of all tarsi with light transverse line on distal end. Anten- niform leg: 17.5 mm, femur, rest is missing. Leg II: 13.8 mm, femur; 18.3 mm, tibia. Leg III: 14.4 mm, femur; 20.9 mm, tibia. Leg IV: 12.0 mm, femur; 18.7 mm (9.0/1. 5/3.4/4.8), tibia; 2.3 mm (1.0/0.4/0. 1/0.8), tarsus. Female Genitalia (Colombia, Cucuta). As in Fig. 126. Variation.— Trinidad specimens present a much shorter spine on the middle face of the pedipalp trochanter, sometimes smaller than some of the larger tubercles on the anfero- external row. Trinidad specimens also might have a smaller, and more difficult to distin- guish, spine on the proximal end of the dorso-inner lateral surface of the pedipalp tarsus. Some Venezuelan specimens present five spines on the trochanter, with an additional spine between spines three and four. Natural History.— In Curasao, Aruba and Trinidad, Phrynus pulchripes has been found inside caves. It has been collected from hollow logs, under rotten bark, under stones, and under coconut husks on a beach. Distribution.— Trinidad (numerous records); Bonaire; Cura9ao; Aruba; Venezuela, Rio Pauguoza and Rio Toro, Caracas and Caripito; and Colombia, Cucuta. Phrynus armasi , new species Figs. 24-29, 118, 125, 127, 128, 174; Map 1 Types.— Male holotype, female paratype from Cueva El Mudo, Catalina de Guines, Prov. La Habana, Cuba, March 1966, deposited in the Academia de Ciencias de Cuba. The specific name is a patronym in honor of Luis F. de Armas, in recognition of his numerous collections of Cuban amblypygids and for generously supplying the collected material for its study. Diagnosis.-Distinct from species that have a small inconspicuous spine on the proxi- mal end of dorso-inner lateral surface of pedipalp tarsus for having a single tooth on the external margin of the basal cheliceral segment, no spine on the anterior trochanter face and for having a small spine at the base of Td-7. The female gonopods are quite distinct from all known species, sclerites being pale yellow-brown. Description.— Male holotype. Carapace light reddish-brown with diffuse yellow mark- ings irradiating from sulcus. Narrow yellow band around carapace edges, wider around posterior half. Light reddish-brown frontal area, darker brown around black median ocular tubercle. Yellowish-brown, uniformly colored femora of legs. Pedipalps and dor- sum of chelicerae reddish-brown. Total length 26.0 mm. Probably not a full-grown adult as female paratype is 31.7 mm long. Carapace. Narrow, well-bilobed anterior edge. Frontal process concealed from above. Distance of median ocular tubercle from anterior edge same as length of tubercle (0.5 mm/0.5 mm). Carapace 9.0 mm long, 12.9 mm wide, 5.3 mm sulcus from anterior edge. Median ocular tubercle 0.6 mm wide. Lateral eyes 4.3 mm from each other, 2.0 mm from anterior edge, 1.5 mm from lateral edge. QUINTERO-THE AMBLYPYGID GENUS PHRYNUS 133 Chelicerae. Dorsal surface of basal segment without distal tubercles. A single tooth on external margin of basal cheliceral segment and a proximal ridge connecting the inner double-pointed tooth. This ridge presents two small lumps along its edge (Fig. 118). Genital Operculum. 3.3 mm long, 6.0 mm wide. Opisthogeminate organ as in Figs. 125 and 127. Abdomen 17.7 mm long. Trichobothria. As in Fig. 174. Ratios: sbc 0.57 and bt 0.38. Pedipalps. Figs. 24-29. Trochanter with four spines. Anterior surface without a setifer- ous tubercle. Femur Fd-3 longer than Fd-2. Fd-4 longer than Fd-6. Fv-1, Fv-2 and Fv-3 each on a separate base. Fv-6 longer than Fv-4. Tibia, Td-6 longer than Td-1 and Td-2. Td-4 longer than Td-2, longer than Td-6. Td-3 and Td-5 are the same size. Tv-4 longer than Tv-6, same size as Tv-1. Basitarsus, Bd-1 very well developed, longer than half the length of Bd-3, Bv-3 longer than Bv-1, Bv-1 shorter than Bd-1. Pedipalp tarsus and post-tarsus completely fused. Tarsus with proximal end of dorso-inner lateral surface with small inconspicuous spine. Femur 10.1 mm long; tibia 11.4 mm long, 2.6 mm wide; basitarsus 5.4 mm long, 2.2 mm wide; tarsus 5.4 mm long. Legs. Second tarsomers of all tarsi without light transverse line on distal end. Antenni- form leg: 33.2 mm, femur; 70.3 mm, tibia; 69.0 mm, tarsus. Leg II: 23.0 mm, femur; 36.3 mm, tibia. Leg III: 25.1 mm, femur; tibia missing. Leg IV: 20.7 mm, femur; 38.7 mm (16.8/4.7/7.2/10.0), tibia; 4.6 mm (2. 0/0. 8/0. 2/1.6), tarsus. Female Genitalia (Cueva El Mudo, Prov. La Habana, Cuba). As in Fig. 128. Natural History.— It has been collected mostly from caves. Distribution. -Cub a: Pinar del Rio and La Habana. Records. -CUBA. Pinar del Rio : Cueva El Indio, Vinales, 2 July 1964, 1 female (P. Alayo, and M. L. Jaume, ACC); Rancho Mundito, June 1947, 1 female (F. Zayas, ACC). La Habana: unnamed cave near Cayo La Rosa, Bauta, 12 March 1972, 1 male (J. Krecek, ACC); Cueva El Indio, Tapaste, 24 March 1956, 1 male, three females (F. Zayas, G. Silva, and Sanchez, ACC); Loma de Candela, 1 July 1950, 2 females (M. L. Jaume, ACC). ST. BART HELEMY £> ANTIGUA 9 Phrynus goesii + Phrynus Tessel Iditus 0 PhrynuS barbadensiS DOM/ % <8 MARTINIQUE STA. LUCIA tfST. VINCENT ^ 4i BARBADOS @ GRENADA Map 2. -Distributions of Phrynus barbadensis, P . gervaisii, P. goesii, P. pulchripes, P. santarensis and P. tessellatus. 134 THE JOURNAL OF ARACHNOLOGY Phrynus goesii Thorell Figs. 42-46, 48, 115, 131, 133, 134, 173; Map 2 Phrynus goesii Thorell 1889:530-533. Syntypes (2 males, 2 females, and one specimen without abdomen) from St. Barthelemy, West Indies, coll. Thorell No. 3015, Dr. Middleship, Vegesack 1836, in the Naturhistoriska Riksmuseum, Stockholm, examined. Designated one female lectotype. See Note below. Tarantula goesii: Pocock 1893:542 (citation only as “species unknown to me”). Tarantula pallasii Pocock 1893:533-534, pi. 40, Fig. 3, male, female. Male holotype from Martinique believed lost from BMNH. Topotypic specimens examined. Pocock, 1894: 278. NEW SYNONY- MY. Tarantula scabra Pocock 1893:540. Male holotype from Monserrat, West Indies, in the BMNH, examined. Pocock 1894:278. NEW SYNONYMY. Tarantula palmata: Kraepelin 1899:242-244 (in part). Kraepelin characterized T. palmata (nec Herbst, 1797) as having Td-4 shorter than Td-2 and erroneously included P. goesii in the list of synonyms, species that has Td-4 longer or same size as Td-2. Note.— Thorell indicated the syntypes were “possidet Mus. Zool. Holmiense,” collec- tions placed at the Mus. Civ. Storia Nat., Genova, which were in large part destroyed in 1970. Fortunately, Thorell syntypes were not destroyed as they had been transferred to Stockholm. Although Thorell (1889) mentioned he had “exempla pauca,” not indicating the exact number of specimens, his description of P. goesii only measures and describes a single specimen. The female designed lectotype, corresponds in all details to the specimen that Thorell described. Diagnosis- Phrynus goesii , P. tessellatus and P. longipes include the largest animals within the genus, body length larger than 32.0 mm. It differs from Phrynus longipes in the number of segments in the antenniform leg: 31 tibial, 68 tarsal for/3, goesii while P. longipes has 33 tibial and 71 tarsal segments . Phrynus goesii is distinct by the presence of a basal proximal spine in Td-6 and Tv-6 and because it is the only species with three teeth on external margin of basal cheliceral segment in which Td-4 is longer than Td-2. Phrynus tessellatus and P. longipes have Td-4 shorter than Td-2. Description.— Female lectotype. Carapace dark reddish-brown, pedipalps darker. Legs reddish-brown without banding. Dorsum of abdomen yellowish-brown without distinct marks, darker brown around muscular impressions. Total length 35.0 mm. Carapace. With narrow, slightly bilobed anterior edge. Frontal process concealed. Distance of median ocular tubercle from anterior edge nearly double length of tubercle (1.2mm/0.7 mm). Carapace 12.8 mm long, 18.4 mm wide, 7.8 mm sulcus from anterior edge. Median ocular tubercle 0.7 mm long, 0.9 mm wide, 1.2 mm from anterior edge. Lateral eyes 7.0 mm from each other, 3.2 mm from anter edge, 3.1 mm from lateral edge. Chelicerae. Anterodorsal surface of basal cheliceral segment with well-developed tu- bercle on outer edge. Three teeth on external margin of basal cheliceral segment. (Fig. 115). Genital Operculum. 4.2 mm long, 8.4 mm wide. Female gonopods as in Fig. 134. Abdomen 23.8 mm long. Trichobothria. As in Fig. 173. Ratios: sbc 0.49, always lower than 0.53. Pedipalps. Figs. 42-46, 48. Trochanter with four spines, no spine at center of anterior surface. Femur Fd-4 very small, less than half the length of Fd-6. Fd-3 longer than Fd-2. Fd-4 a denticle. Fv-1 and Fv-2 on a common base distinct from Fv-3. Tibia, Td-6 with a well-developed proximal basal spine. Td-4 longer than Td-2 and Td-6. Td-1 and Td-7 about the same size. Tv-6 with a proximal basal spine. Tv-4 and Tv-6 are the same size. QUINTERO-THE AMBLYPYGID GENUS PHYRNUS 135 Three distal well-developed setiferous tubercles. Basitarsus Bd-1 distinct but shorter than Bd-3. Bv-1 slightly shorter than Bv-3. Tarsus dorso-inner lateral surface with a small inconspicuous spine located proximally; tarsus and post-tarsus fused. Femur 13.5 long; tibia 14.2 mm long, 3.8 mm wide; basitarsus 7.1 mm long; tarsus 7.3 mm long. Legs. Second tarsomere with incomplete membranous line on each side. Antenniform leg: 28.2 mm, femur; 46.0 mm, tibia; 52.0 mm, tarsus. Leg II: 22.0 mm, femur; 33.9 mm, tibia. Leg III: 22.3 mm, femur; 35.0 mm, tibia. Leg IV: 18.2 mm, femur; 32.2 mm (14.8/3.3/5.3/8.8), tibia; 4.8 mm (1.9/0.7/0.4/1.8), tarsus. Male paratype (St. Barthelemy). Genitalia as in Figs. 131 and 133. Variation. -Frontal margin pronouncedly emarginated and with darker coloration in St. Kitt specimens. Largest size animals in St. Barthelemy. Natural history.— It has been collected from a cave in St. Martin and among the ruins of a sugar mill in Nevis Island. Distribution. -Anguilla, St. Martin, Saba, St. Eustatius, St. Kitts, Nevis, Antigua, Montserrat, Doninica and Martinique. Phrynus tessellatus (Pocock) Figs. 47, 49-53, 114, 149, 151, 170; Map 2 Tarantula tessellata Pocock 1893:531-3, pi. 40, Fig. 2, male and female. Male holotype from St. Vincent, West Indies, in the BMNH, examined. Specimen in poor condition, carapace fractured in two pieces, held by underlying muscles, sternites and gnathocoxae crushed. Pocock 1894:278. Tarantula spinimana Pocock 1893:534-6, pi. 40, Fig. 4. Male holotype discolored and in poor condi- tion, from Haiti, in BMNH, examined. Locality probably erroneous. Pocock 1894:278. NEW SYNONYMY. Neophrynus palmatus: Kraepelin 1895:30-34 (in part). Admetus palmatus: Simon 1897:890 (record only). Tarantula palmata: Kraepelin 1899:242-244 (in part). Phrynus tesselatus (sic): Mello-Leitao 1931:41-44. Diagnosis. —Phrynus tessellatus is closely related to Phrynus goesii and P. longipes. They are species of large size (larger than 32.0 mm long) that have three teeth on the external margin of basal cheliceral segment and an inconspicuous small spine on the dorso-inner lateral surface of tarsus. Phrynus goesii and P. longipes have their frontal process concealed while in P. tessellatus the frontal process stands out long and pointed between the chelicerae basis. Most individuals of Phrynus tessellatus have a very distinct checkerboard pattern on their abdominal tergites. Phrynus tessellatus differs from Phrynus longipes in having the femur of the antenniform leg 2.3 to 2.7 times longer than the median prosomal length, the sbc trichobothria higher than 0.53 and Bv-1 longer than Bv-3. Phrynus longipes has the femur of antenniform leg 2.8 to 3.8 times longer than medial prosomal length, sbc trichobothria lower than 0.53 and Bv-1 shorter or same size as Bv-3. Description.-Male holotype. Carapace and pedipalps dark reddish-brown. Legs reddish-brown with two ill-defined flavous bands on femur IV. Dorsum of abdomen with a checkerboard appearance of rectangles and squares either reddish-brown or yellow, each tergite bearing approximately 10 spots, 5 anterior and 5 posterior, alternating dark and light spots. Total length 25.8 mm. It is not a full-grown individual, but Pocock selected for description “the largest male that was least fractured” of the eight specimens he had from St. Vincent. The largest specimen that I have seen of this species was 34.0 mm long. 136 THE JOURNAL OF ARACHNOLOGY Carapace. With narrow anterior edge, distinctly emarginated and conspicuously den- tate. Long and pointed frontal process standing out between the base of chelicerae. Distance of median ocular tubercle from anterior edge more than double length of tuber- cle (1.9 mm/0.8 mm). Carapace 10.8 mm long, 17.4 mm wide, 6.2 mm sulcus from anterior edge. Median ocular tubercle 1.1 mm wide. Lateral eyes 5.2 mm from each other, 1.9 mm from anterior edge, 2.4 mm from lateral edge. Chelicerae. Anterodorsal surface of basal cheliceral segment with well-developed tuber- cle on outer edge. Three teeth on external margin of basal cheliceral segment (Fig. 114). Gential Operculum. 4.3 mm long, 7.1 mm wide. Male genitalia as in Figs. 149 and 151. Abdomen 17.9 mm long. Trichobothria. As in Fig. 170. Ratios: sbc 0.57, always higher than 0.53. Pedipalps. Figs. 47, 49-53. Trochanter with four spines, tubercle at center of anterior surface shorter than largest tubercle on antero-external row. Femur Fd-3 longer than Fd-2, Fd-4 very small. Fv-4 about same size as Fv-7, and more than half the length of Fv-6. Fv-1 and Fv-2 on a common base distinct from Fv-3. Tibia, Td-6 without proximal basal spine. Td-4 shorter than Td-2, longer than Td-6. Td-1 longer than Td-7. Tv-6 without proximal basal spine. Tv-4 longer than Tv-6. Basitarsus Bd-1 well developed, more than half the length of Bd-3. Bv-1 longer than Bv-3. Tarsus dorso-inner lateral surface with a small inconspicuous spine located proximally; tarsus and post-tarsus fused. Femur 12.0 mm long; tibia 13.5 mm long, 3.4 mm wide; basitarsus 5.9 mm long; tarsus 6.0 mm long. Figs. 42-5 3. -Pedipalps: Phrynus goesii Thorell, female lectotype, 42-46, 48; Phrynus tessellatus (Pocock), male holotype, 47, 49-53. Figs. 42, 49 = femur, dorsal view; 45, 52 = femur, ventral view; 43, 50 = tibia, dorsal view; 46, 53 = tibia, ventral view; 44, 51 = basitarsus and tarsus, inner lateral view; 47, 48 = trochanter, anterodorsal view. QUINTERO-THE AMBLYPYGID GENU S PHR YNUS 137 Legs. Second tarsomere with incomplete membranous line not reaching ventral edge on each side. Antenniform leg: 28.8 mm, femur; tibia and femur missing. Leg II: 20.6 mm, femur; 25.9 mm, tibia. Leg III: 20.8 mm, femur; 28.8 mm, tibia. Leg IV: 17.6 mm, femur; 28.4 mm (13.1/2.0/4.9/7.4), tibia; 4.2 mm(1.6/0.8/0.2/1.6), tarsus. Variation.— Common differences in the pattern on the abdominal tergites. Natural History.— In St. Vincent, specimens have been taken at altitudes of 150 to 1,000 ft. (45-305 m) by streams (under the bark of stumps) and immatures under stones. In Grenada, one immature was collected under a stone near sea level. Distribution. -St. Vincent, St. Lucia and Grenada. Phrynus longipes (Pocock) Figs. 36-41, 116, 129, 130, 132, 172; Map 1 Tarantula longipes Pocock, 1893:536-7, pi. 40, Fig. 5, male, female. Male holotype from Haiti, in the BMNH, examined. Pocock, 1894:277. Tarantula thorellii Pocock, 1894:282-3, pi. 7, Fig. 7, male. Male holotype with no collecting informa- tion in the BMNH, examined. NEW SYNONYMY. Neophrynus palmatus: Kraepelin 1895:30-34 (in part). Tarantula palmata : Kraepelin 1899:242-244 (in part). Phrynus longipes'. Mello-Leitao 1931:40-42. Diagnosis— Phrynus longipes and P. pulchripes are closely associated by their similar spination dorsal of the pedipalp tibia. It differs fromP. pulchripes by having three teeth, instead of two, on the external margin of the basal cheliceral segment and by lacking the spine on the anterior face of the pedipalp trochanter. It differs from other species with a small inconspicuous spine on the proximal end of the dorso-inner lateral surface of the pedipalp tarsus by its very long femur of the antenniform legs that reaches from 2.8 to 3.8 times the median prosomal length, by having the frontal process concealed and Bv-1 shorter than Bv-3. Description.— Male holotype. Carapace and pedipalps light reddish-brown Carapace with two diffuse flavous stripes each at its posterior lateral edges. Legs and abdomen lighter colored. Distinct banding on the femora of the walking legs. Abdomen variegated with brown and yellow. Total length 24.0 mm . Carapace. With wide, nearly straight anterior edge, with small denticles slightly larger at the sides. Frontal process concealed from above. Carapace furnished with only a few coarse granules. Very distinct frontal area, convex in front of lateral eyes. Distance of median ocular tubercle from anterior edge one-third length of tubercle (0.3 mm/0.9 mm). Carapace 9.0 mm long, 13.3 mm wide, 5.7 mm sulcus from anterior edge. Median ocular tubercle 1.0 mm wide. Lateral eyes 4.3 mm from each other, 1.6 mm from anterior edge, 1.5 mm from lateral edge. Chelicerae. Anterodorsal surface of basal cheliceral segment with two tubercles at the distal border, of which the external is the larger. Both tubercles are small. Three teeth on external margin of basal cheliceral segment (Fig. 116). Genital Operculum. 3.4 mm long, 5.8 mm wide. Opisthogeminate organ as in Figs. 129 and 130. Abdomen 15.5 mm long. Trichobothria. As in Fig. 172. Ratios: sbc 0.51, bt 0.57. Pedipalps. Figs. 36-41. Trochanter with four spines. A small setiferous tubercle at the center of anterior surface, smaller than larger tubercles on antero-external row. Femur 138 THE JOURNAL OF ARACHNOLOGY and tibia studded, but not thickly, with small granules. Femur, Fd-2 longer than Fd-3. Fd-4 small, shorter than Fd-5 and Fd-6. Fv-1 and Fv-2 on a common base separate from Fv-3. Fv-4 one-fourth the length of Fv-6. Fv-3 longer than Fv-6. Tibia, Td-6 longer than Td-1 but shorter than Td-2. Td-4 shorter than Td-2 and longer than Td-6. Td-3 longer than Td-5. Tv-4 longer and thinner than Tv-6. Tv-4 and Tv-1 are the same size. Basitarsus, Bd-1 long, longer than half the length of Bd-3. Bv-3 longer than Bv-1. Bv-1 and Bd-1 are the same size. Pedipalp tarsus and post-tarsus completely fused. Tarsus with proximal end of dorso-inner lateral surface with small, inconspicuous spine. Femur 1 1 .0 mm long; tibia 1 1.3 mm long, 2.4 mm wide; basitarsus 5.9 mm long, 2.4 mm wide; tarsus, 5.5 mm long. Legs. Antenniform leg: 33.9 mm, femur; rest is missing. Leg II: 21.6 mm, femur; 30.0 mm, tibia. Leg III: 24.2 mm, femur; tibia lost. Leg IV: 20.8 mm, femur; 36.0 (15.1/4.7/6.5/9.7), tibia, all tarsi missing. Female Genitalia (Puerto Rico, Cueva de los Alfaros). As in Fig. 132. Natural History.— In the Dominican Republic and Puerto Rico, Phrynus longipes has been collected from caves. In St. John, Virgin Is., one gravid female was collected from under a rock in a moist forest and two males from a sunny dry to very dry area. In Tortola Is. one male was found at 1,200 ft. in an old water cistern on peak above town. Distribution.— Haiti, Dominican Republic, Puerto Rico, and Virgin Islands: St. John, St. Croix, St. Thomas, and Tortola Island. Phrynus damonidaensis , new species Figs. 54-59, 122, 135, 136, 137, 164; Map 1 Types.— Female holotype from Uvero, El Cobre, Sierra Maestra, Prov. Oriente, Cuba (L. de Armas), 25 May 1972; 3 male, 2 female paratypes from Cayo Dama, Chivirico, El Cobre, under rocks (L. de Armas), 24 May 1972; deposited in the Academia de Ciencias de Cuba. The specific name is a noun after the name of the family Damonidae, having in common only three tibial segments in leg IV. Diagnosis.— Phrynus damonidaensis can be easily recognized by its peculiar tibial seg- mentation, having only three tibial IV segments instead of four segments. The only exception is Phrynus santarensis whose exact number of tibial IV segments is not known with certainty because of the asymmetry present in the single known specimen, which presents three segments on left tibia IV and only two segments on right tibia IV. A similar asymmetry was found in the female holotype of P. damonidaenisis, but all other known specimens present three tibial IV segments. Phrynus damonidaensis is not a member of the Damonidae, 'as redefined by Quintero (1976) for the following reasons: 1. Its inner proximal double-pointed tooth of the basal cheliceral segment has its distal cusp distinctly larger. In all members of the Damonidae the proximal cusp is the larger of the two cusps. 2. It lacks the baso-caudal row of trichobothria on the distitibia of leg IV. This row of trichobothria is present in all members of the Damonidae. Description.— Female holotype. Carapace light reddish-brown with diffuse yellow markings irradiating from sulcus. Narrow white band around carapace edges, wider a- round posterior half and on frontal margin. Yellow tubercles scattered on posterior half of carapace and dorsum of abdomen. Darker reddish-brown behind lateral eyes and QUINTERO- THE AMBLYPYGID GENUS PHRYNUS 139 yellow lines beside lateral eyes. Dark reddish-brown pedipalps. Yellowish-brown, uniform- ly colored femora of legs. Dorsal surface of abdomen approximately same color as legs, with a variegated pattern, lighter around muscular impressions. Total length 17.6 mm. Carapace. Wide, nearly straight anterior edge, with uneven size tubercles, larger at the sides. Frontal process concealed from above. Distance of median ocular tubercle from anterior edge more than double length of tubercle (0.8 mm/0.3 mm). Carapace 6.7 mm long, 9.5 mm wide, 4.0 mm sulcus from anterior edge. Median ocular tubercle 0.6 mm wide. Lateral eyes 4.3 mm from each other, 1.6 mm from anterior edge, 1.0 mm from lateral edge. Chelicerae. Four teeth on external margin of basal cheliceral segment. Of these teeth, the distal one is located proximally on the lower third of the large ventral, distal tooth. The other three teeth insert on a common base. The proximal one is very small (see comments under Variation) (Fig. 122). Genital Operculum. 2.3 mm long, 4.3 mm wide. Female gonopods as in Fig. 137. Abdomen 10.8 mm long. Trichobothria. As in Fig. 164. Ratios: sbc 0.44, bt 0.37. Pedipalps. Figs. 54-59. Trochanter with four spines. Center of anterior surface without a spine. Femur with numerous coarse granules on its dorsal surface, sparsely granulated ventrally. Femur, Fd-2 longer than Fd-3. Fd-4 longer than Fd-5 and Fd-6. Fv-1, Fv-2 and Fv-3 present on a common base. Fv-4 a small tubercle. Fv-3 longer than Fv-6. Tibia, Td-6 Figs. 5 4-65. -Pedipalps of female holotypes: Phrynus damonidaensis, new species, 54-59 \ Phrynus marginemaculatus C. L. Koch, 60-65. Figs. 54, 61 = femur, dorsal view; 57, 64 = femur, ventral view; 55, 62 = tibia, dorsal view; 58, 65 = tibia, ventral view; 56, 63 = basitarsus and tarsus, inner lateral view; 59, 60 = trochanter, anterodorsal view. 140 THE JOURNAL OF ARACHNOLOGY same size as Td-1 and less than half the length of Td-2. Td-4 shorter than Td-2 and longer than Td-6. Td-3 longer than Td-5. Tv-4 longer than Tv-6. Tv-4 shorter than Tv-1. Basi- tarsus, Bd-1 short, about one-third the length of Bd-3. Bv-3 shorter than Bv-1 and reduced to an inconspicuous tubercle. Bv-1 shorter than Bd-1. Pedipalp tarsus and post-tarsus completely fused. Tarsus without small spine on proximal end of dorso-inner lateral surface. Femur 6.5 mm long; tibia 7.2 mm long, 2.2 mm wide; basitarsus 3.2 mm long; 1 .9 mm wide; tarsus 3.4 mm long. Legs. Second tarsomere of all tarsi with light transverse complete line on distal end. The left side presents a curious asymmetry, the tibia IV divided into only two segments while the right tibia IV has the normal number of segments for this species, three. Antenniform leg: 1.42 mm, femur; 27.0 mm, tibia; 29.0 mm, tarsus. Leg II: 10.0 mm, femur; 14.0 mm, tibia. Leg III: 10.9 mm, femur; 16.0 mm, tibia. Leg IV: 9.3 mm, femur; 15.2 mm (7.1/3.2/4.9), tibia; 2.5 mm (1.1/0.5/0.2/0.7), tarsus. Male Genitalia (Uvero, El Cobre, Prov. Oriente, Cuba). As in Figs. 135 and 136. Variation.— Adult specimens from Swan Islands, western Caribbean Sea, differ from Cuban specimens in having a darker general appearance. They also differ in having the three proximal teeth on the external margin of the basal cheliceral segment of even size, instead of having the proximal tooth very reduced in size. Two males from Cayo Dama, Prov. Oriente, Cuba, present the fifth abdominal sternite divided with a membranous line along the middle. Natural History. -It has been collected from under rocks in Oriente, Cuba. The num- ber of embryos in six females carrying egg cases was 7, 8, 12, 18, 39 and 50 (mean of 22 Figs. 66-77.-Pedipalps: Phrynus levii new species, male holotype, 66-7 l;Phrynus santarensis (Po- cock), female holotype, 72-77. Figs. 66, 73 = femur, dorsal view; 69, 76 = femur, ventral view; 67, 74 = tibia, dorsal view; 70, 77 = tibia, ventral view; 68, 75 = basitarsus and tarsus, inner lateral view; 71, 72 = trochanter, anterodorsal view. QUINTERO-THE AMBLYPYGID GENUS PHYRNUS 141 Distribution.— Cuba, Pinar del Rio and Oriente; Honduras, Swan Islands. Records.-CUBA. Oriente : Sierra Maestra, Uvero, El Cobre, 25 May 1972, 2 males (L. de Armas, ACC). Baitiquiri, Guantanamo, under rock, 23 October 1971, 1 male, juvenile (J. de Cruz, ACC). Cayo Dama, Chivirico, El Cobre, under rocks, 24 May 1972, 3 males, 2 females (L. de Armas, ACC). Morro, Santiago de Cuba, 22 May 1972, 1 male (L. de Armas, ACC). Finca El Curoe, S. W. Guan- tanamo, 26 November 1950, 1 male (P. Alayo and S. de Torre, ACC). Laguna Baconeo, W. Guan- tanamo, 21 August 1966, 1 male, 3 females (M. L. Jaume, ACC). Margins Siboney R., El Caney, 7 November 1971, 2 males, 1 female (L. de Armas, ACC). Siboney El Caney, 5 November 1971, 2 males, 2 females (L. de Armas, ACC). Juragua, El Caney, 4 November 1971, 3 males, 3 females (L. de Armas, ACC). Pinar del Rio : El Veral, Pen. Guanahacabibes, 28 August 1971, 1 male, 2 females (L. de Armas, ACC). HONDURAS: Swan Islands, juvenile, 14 November 1937 (J. S. Colman and M. Y. Rosaura, BMNH), 1 male, 6 females, 17-19 April 1913 (Geo. Nelson, MCZ). Phrynus marginemaculatus C. L. Koch Figs. 60-65, 120, 141, 142, 144, 166; Map 1 Phrynus marginemaculatus C. L. Koch 1841:6-8, Fig. 597. Female holotype form West Indies, no other information, in the BMNH, examined. Admetus marginemaculatus: C. L. Koch 1850:81. Phrynus pallasii Blanchard 1852-64:170, pis. 10 bis and 11. Blanchard does not appear to have left specimens in the Paris Museum, where he worked. He indicated that his specimens came from different localities in the West Indies. Tarantula keyserlingii Pocock 1893:539-540, pi. 40, Fig. 7. The holotype came from an unknown locality, appears to be lost from the BMNH. Male paratype from Cuba, in the BMNH, examined. Tarantula latifrons Pocock 1893:537-9, pi. 40, Fig. 6. Female holotype from Haiti, in the BMNH, examined. Pocock 1894:278. Tarantula marginemaculata: Pocock 1893:541. Pocock indicated that this species was unknown to him, thus he was unable to place it in his key to species. Kraepelin 1899:245, Fig. 89. Muma 1967:24-25, Figs. 18-19. Weygoldt 1969:338-360; 1970:58-85. Neophrynus marginemaculatus: Kraepelin 1895:34-36. Phrynus keyserlingi (sic) : Mello-LeitSo 1 9 3 1 : 42 . Diagnosis.— Phrynus marginemaculatus, P. levii and P. damonidaensis are the only spe- cies with 27 tibial segments in the antenniform leg. Phrynus parvulus has 25 tibial seg- ments and all the other species of Phrynus have 29 or more tibial segments in the antenniform leg. Phrynus marginemaculatus differs from P. damonidaensis in having four tibial segments on leg IV instead of three, they are darker colored animals and have two teeth instead of four on the external margin of the basal cheliceral segment. It differs from P. levii in being a smaller, darker colored animal with different ornamentation of the carapace. It has two teeth instead of three on the basal cheliceral setment and has Fv-1 and Fv-2 with a common base, distinct from Fv-3. Description.— Female holotype. Carapace and pedipalps dark reddish-brown. Femora of legs dark brown, uniformly colored. Dorsal of cheliceral basal segment lighter than frontal area. Carapace with two distinct pale yellow spots on each posterior ectal angle, and a yellow line beside each lateral ocular cluster. Carapace and abdomen sprinkled with light tubercles. Each abdominal tergite has one pale yellowish large spot medial to the two dark brown muscular impressions. Total length 12.5 mm. It does not appear to be a full-grown adult, adults reach a maximum size of 18.0 mm. Carapace. Narrow, well-bilobed anterior border, with very small, uneven size tubercles. Frontal process concealed from above. Distance of median ocular tubercle from anterior 142 THE JOURNAL OF ARACHNOLOGY edge slightly larger than length of tubercle (0.5 mm/0.4 mm). Carapace 5.7 mm long, 9.0 mm wide, 3.4 mm sulcus from anterior edge. Median ocular tubercle 0.6 mm wide. Lateral eyes 3.7 mm from each other, 1.6 mm from anterior edge, 0.6 mm from lateral edge. Chelicerae. Two teeth on external margin of basal cheliceral segment (Fig. 120). Genital Operculum. 2.0 mm long, 4.0 mm wide. Female gonopods as in Fig. 144. Abdomen 8.3 mm long. Trichobothria. As in Fig. 166. Ratios: sbc 0.40, bt 0.25. Pedipalps. Figs. 60-65. Trochanter with four spines. Center of anterior surface without a spine. Femur, Fd-2 longer than Fd-3, Fd-4 longer than Fd-5. Only five spines present. Fv-1 and Fv-2 with a common base separate from Fv-3. Fv-4 obsolete. Fv-3 longer than Fv-6. Tibia, Td-6 longer than Td-1. Td-2 longer than Td-6. Td-4 shorter than Td-2 and longer than Td-6. Td-3 same size as Td-5. Tv-4 about same size as Tv-6 and shorter than Tv-1. Tv-2 longer than Tv-5. Basitarsus, Bd-1 inconspicuous, a basal appendage to Bd-2. Bv-3 and Bv-1 both inconspicuous. Pedipalp tarsus and post-tarsus completely fused. Tarsus without small spine on proximal end of dorso-inner lateral surface. Femur 5.2 mm long; tibia 6.0 mm long, 2.0 mm wide; basitarsus 2.8 mm long, 1.7 mm wide; tarsus 3.0 mm long. Legs. Second tarsomere of all tarsi with light transverse complete line on distal end. Antenniform leg: 12.0 mm, femur; 24.5 mm, tibia; 30.0 mm, tarsus. Leg II: 8.8 mm, femur; 12.2 mm, tibia. Leg III: 9.0 mm, femur; 12.9 mm, tibia. Leg IV: 8.0 mm, femur; 1 1.5 mm (5.1/0.8/2.4/4.2), tibia; 2.0 mm (0.9/0. 3/0. 1/0.7), tarsus. Male Genitalia (Isla de Pinos, Cuba). As in Figs. 141 and 142. Variation.— The species derives its name from the pale yellowish spots on each poster- ior ectal angle of the carapace. Unfortunately, this character is also common among other species of Phrynus and variable in P. marginemaculatus . The marginal spots on the cara- pace could even be absent. This condition is frequently found among specimens from Isla de Pinos, Cuba. Natural History.— Muma (1967) states that, in southern Florida , P. marginemaculatus is found commonly under boards, logs, and trash on the ground, under the bark of dead trees, and on and in houses. Muma maintained living specimens in the laboratory for longer than one year, using termites as food. In six preserved females, the number of embryos carried ranged from 17 to 36 with a mean of 24. Weygoldt (1969) studied the behavior of specimens from Big Pine Key and Key Largo, Florida, and described the mating and ritualistic interactions in this species. He observed that the molting occurs during the night while they hang in an inverted position from a rock. It lasts several hours and is finished in the morning. The species has been collected in timber, among cedar shingles, in Jamaica, and one gravid female under rotten log in New Providence Is., Bahamas. Distribution.— USA. Southern Florida, as far north as Sunniland, Collier County in the center of the state, Martin County on the east coast, and Punta Gorda on the west coast (Muma, 1967). Monrow County up to Dry Tortugas. Bahamas: Six Hills Cay, Abaco, Andros Is., Man O’War Cay, Crooked Is., Great Inagua Is., South Bimini Is., Exuma Is. (Bitter Guana Cay), New Providence Is., Long Is., San Salvador Is. (Watling), Rum Cay, Long Cay (S. of S. Caicos Is.), Turks Is., and Eleuthera Island. Cuba: collected from all six provinces. Jamaica, Haiti, and Dominican Republic. QUINTERO-THE AMBLYPYGID GENUS PHRUNUS 143 Phrynus levii , new species Figs. 66-71, 121, 138-140, 161; Map 1 Types.— Male holotype, female paratype from Providence Cave, Montego Bay, Jamaica, 5 March 1911, deposited in the BMNH. The specific name is a patronym in honor of Dr. Herbert W. Levi in recognition of the encouragement he gave me to continue work on the taxomony of Phrynus , and his contributions in the field of arachnology. Diagnosis.— Phrynus levii belongs to the group of species which lack the small spine on the proximal end of the dorso-inner lateral surface of the pedipalp tarsus and have Td-4 shorter than Td-2. It appears closely related to P. marginemaculatus , differing in having three teeth instead of one on the basal cheliceral segment. They are larger animals, 24.0 mm maximum total length found, while the largest P. marginemaculatus measured was 18.0 mm. Phrynus levii, appears lighter in general body coloration, particularly the abdo- men with a plain yellowish-brown look, while P. marginemaculatus has it variegated and darker. In P. levii the chelicarae are always darker than the frontal area of the carapace while in P. marginemaculatus they are lighter, rarely the same color as the fontal area. Description. -Male holotype. Carapace light reddish-brown with diffuse yellow mark- ings irradiating from sulcus. Pedipalps and dorsal of cheliceral basal segment of darker reddish-brown. Frontal area clearly lighter than chelicerae. Yellow lines beside lateral eyes. Wide yellow band around posterior half of carapace edges. Yellowish-brown, uni- formly colored femora of legs. Total length 23.5 mm. Carapace. Anterior edge well bilobed, narrow and evenly denticulated with short tu- bercles. Frontal process concealed from above. Distance of median ocular tubercle from anterior edge nearly double length of tubercle (0.9 mm/0.5 mm). Carapace 8.9 mm long, 12.9 mm wide, 5.2 mm sulcus from anterior edge. Median ocular tubercle 0.9 mm wide. Lateral eyes 5.4 mm from each other, 2.0 mm from anterior edge, 1.3 mm from lateral edge. Chelicerae. Dorsal surface of basal segment with medial, distal tubercle enlarged. Three teeth on external margin of basal cheliceral segment (Fig. 121). Genital Operculum. 3.7 mm long, 6.2 mm wide. Opisthogeminate organ as in Figs. 138 and 139. Abdomen 15.0 mm long. Trichobothria. As in Fig. 161. Ratios: sbc 0.45 and bt 0.37. Pedipalps. Figs. 66-71. Trochanter with four spines. Anterior surface without a spine. Femur, Fd-3 longer than Fd-2. Fd-4 shorter than Fd-5 and Fd-6. Fv-1, Fv-2 and Fv-3 without a distinct common base. In place of Fv-4, two short convergent spines. Fv-3 longer than Fv-6. Tibia Td-6 three times the length of Td-1 and more than half the length of Td-2. Td-4 shorter than Td-2 and longer than Td-6. Td-3 longer than Td-5. Tv-4 slightly longer than Tv-6 and Tv-1. Basitarsus, Bd-1 short, inconspicuous, about one-fifth the length of Bd-3. Bv-3 small, shorter than Bv-1. Bv-1 longer than Bd-1. Pedipalp, tarsus and post-tarsus completely fused. Tarsus without small spine on proximal end of dorso- inner lateral surface. Femur 9.0 mm long; tibia 10.4 mm long, 3.0 mm wide; basitarsus 4.3 mm long, 2.8 mm wide; tarsus 5.2 mm long. Legs. Second tarsomere of all tarsi with light transverse complete line on distal end. Antenniform leg: 26.0 mm, femur; rest is broken. Leg II: 16.3 mm, femur; 24.2 mm, tibia. Leg III: 17.0 mm, femur; 26.3 mm, tibia. Leg IV: 14.8 mm, femur; 30.4 mm (15.1/2.1/5.0/8.2), tibia; 4.2 mm (1.9/0. 8/0. 3/1. 2), tarsus. Female Genitalia (Providence Cave, Montego Bay, Jamaica). As in Fig. 140. Natural History.— It was collected from a cave in Montego Bay, Jamaica. 144 THE JOURNAL OF ARACHNOLOGY Distribution.— Jamaica and Cuba. Records. -CUBA: Las Villas, Rancho Luna, Cienfuegos. 10 March 1955, 2 males (M. Goenaga, ACC). JAMAICA: Providence Cave, Montego Bay, 5 March 1911, 2 females (BMNH). Phrynus santarensis (Pocock) Figs. 72-77, 112, 150, 165; Map 2 Tarantula santarensis Pocock 1894:284-5, 1 female. Female holotype from Santarem, Brazil, in the BMNH, examined. Admetus santarensis: Pocock 1897:358-9. Neophyrynus palmatus barbadensis'. Kraepelin 1895:30-34 (in part). Tarantula palmata santarensis: Kraepelin 1899:244. Phrynus santarensis: Mello-Leitao 1931:43. Diagnosis.— Phrynus santarensis presents a curious asymmetry in the segmentation of the tibia of leg IV. The right tibia has only two segments while the left one has three segments. The only specimen seen of the species is the holotype, thus it is not possible at the present to know with certainty what is the normal number of segments of tibia IV without examining additional specimens. If the number of segments turns out to be three instead of four, it will be the second species known with only three segments in tibia IV, the other being Phrynus damonidaensis. If the number of segments is two, P. santarensis will be the only species within the Phrynidae with that number of segments. Not being able to solve this problem I have not used this character for the diagnosis of the species. Phrynus santarensis can be recognized by the five spines on the edges of the pedipalp trochanter, its lack of the inconspicuous spine on the proximal end of the dorso-inner lateral surface of the pedipalp tarsus, by having Td-4 shorter than Td-2 and by the well-developed spine at center of anterior face of the pedipalp trochanter. Description.— Female holotype. Carapace, pedipalps and legs dark reddish-brown. Cara- pace with diffuse marginal spots on its posterior edges. Abdomen with a ferruginous coloration, with three faintly defined fuscous patches on each tergite, one being median and the others lateral. Total length, 20.0 mm. Carapace. With narrow, slightly bilobed anterior edge with short, even setiferous tuber- cles. Short frontal process partially visible from above. Poorly defined frontal area. Cara- pace very broad and narrow. Distance of median ocular tubercle from anterior edge more than half of length of tubercle (0.4 mm/0.6 mm). Carapace 7.0 mm long, 1 1 .8 mm wide, 4.0 mm sulcus from anterior edge. Median ocular tubercle 0.9 mm wide. Lateral eyes 3.9 mm from each other, 1.5 mm from anterior edge, 1.5 mm from lateral edge. Chelicerae. Anterodorsal surface of basal cheliceral segment without distal tubercles. One well-developed tooth and a square ridge on external margin of basal cheliceral seg- ment. Square ridge located between external tooth and the inner double-pointed proxi- mal tooth (Fig. 112). Genital Operculum. 2.4 mm long, 5.0 mm wide. Female gonopods as in Fig. 150. Abdomen 12.0 mm long. Trichobothria. As in Fig. 165. Ratios: sbc 0.44, bt 0.43. Pedipalps. Figs 72-77. Robust in appearance. Trochanter with five spines, one addi- tional spine between spines 1 and 2. Femur Fd-2 same size as Fd-3, Fd-4 longer than Fd-5 and Fd-6. Fv-1, Fv-2 and Fv-3 present a common base. Fv-4 inconspicuous, reduced to a small tubercle, Fv-3 longer than Fv-6. Tibia, medial face all granular. Td-6 longer than QUINTERO-THE AMBLYPYGID GENUS PHRYNUS 145 Td-1 but shorter than Td-2. Td-4 shorter than Td-2 and longer than Td-6. Td-3 longer than Td-5. Tv-4 longer than Tv-6. Tv-4 same size as Tv-1. Tv-2 and Tv-5 are the same size. Basitarsus, Bd-1 long, more than half the length of Bd-3, Bv-3 longer than Bv-1. Bv-1 shorter than Bd-1. Pedipalp tarsus and post-tarsus completely fused. Tarsus without small spine on proximal end of dorso-inner lateral surface. Femur 6.0 mm long; tibia 6.9 mm long, 2.8 mm wide; basitarsus 3.9 mm long, 2.1 mm wide; tarsus 3.6 mm long. Legs. Second tarsomere of all tarsi with light transverse complete line on distal end. The right side presents the tibia IV divided into only two segments while the left tibia IV is divided into three segments. It is not known whether this is a normal character for the species or due to an abnormality in development, as suggested by Pocock (1894, p. 285). Antenniform leg: 16.0 mm, femur; 30.2 mm, tibia; tarsus missing. Leg II: 11.2 mm, femur; 15.7 mm, tibia. Leg III: 13.0 mm, femur; 18.0 mm, tibia. Leg IV: 10.0 mm, femur; 16.1 mm, femur; 18.0 mm, tibia. Leg IV: 10.0 mm, femur; 16.1 mm (7.9/2. 7/5. 5), tibia; 2.5 mm (1.0/0. 5/0. 2/0. 8), tarsus. Male Genitalia. Not known. Natural History.— In 1897, Pocock reported “many specimens were taken at Santarem (Brazil), one in a house, a few in the forest, and many from a termite’s nest upon the campus.” I was not able to find these specimens at the BMNH. Distribution.— Only known from the type locality, Santarem, Brazil. Phrynus barbademis (Pocock) Figs. 78-83, 111, 153, 169; Map 2 Tarantula barbademis Pocock 1893:529-530, pi. 40, Fig. 1, 1 male, 1 female. Male holotype from Barbados, West Indies, in the BMNH, examined. Specimen in poor condition, legs and pedipalps fractured. Pocock 1894:278. Neophrynus palmatus barbademis : Kraepelin 1895:30-34 (in part). Tarantula palmata barbademis : Kraepelin 1899:244 (in part). Phrynus barbademis'. Pocock 1902a:51, pi. 10, Fig. 6; Mello-Leitao 1931:41. Diagnosis.— It is most closely related to Phrynus gervaisii. Phrynus barbadensis can be recognized by its raised, distinctly delimited, darker frontal area and by the pale, yellowish-brown abdomen with scarcely a trace of pattern. Frontal process broad, visible from above although vertically positioned. Description, Male holotype.— Carapace and pedipalps dark reddish-brown. Carapace with a flavous posterolateral border. Legs paler than pedipalps, with very faintly defined flavous spots dorsal of femora. Abdomen pale yellowish-brown, with scarcely a trace of pattern. Total length 19.0 mm. Carapace. With wide, distinctly bilobed anterior edge, evenly denticulate. Broad fron- tal process partially visible from above and vertically positioned. Raised, distinctly delim- ited frontal area, darker in coloration. Distance of median ocular tubercle from anterior edge nearly half the length of tubercle (0.3 mm/0.5 mm). Carapace 7 0 mm long, 11.1 mm wide, 4.3 mm sulcus from anterior edge. Median ocular tubercle 0.8 mm wide. Lateral eyes 4.5 mm from each other, 1.5 mm from anterior edge, 0.7 mm from lateral edge. Chelicerae. Scarcely granular above, and without an enlarged terminal tubercle. One well -developed tooth and a blunt ridge on external margin of basal cheliceral segment. Blunt ridge located between external tooth and the inner double-pointed proximal tooth (Fig 111). 146 THE JOURNAL OF ARACHNOLOGY Gential Operculum. 3.3 mm long, 5.3 mm wide. I was not able to illustrate the male genitalia because it was poorly preserved and deteriorated. Abdomen 1 1 .3 mm long. Trichobothria. As in Fig. 169. Ratios: sbc 0.45, bt 0.40. Pedipalps. Figs. 78-83. Robust in appearance. Trochanter with four spines, and a well-developed spine at center of anterior surface. Femur, Fd-2 same size as Fd-3. Fd-4 small, one-third the length of Fd-5, and half the length of Fd-6. Fv-1, Fv-2 and Fv-3 on a distinctly raised common base. Fv-4 more than half the length of Fv-6. Fv-3 same size as Fv-5. Fv-3 larger than Fv-6. A short spine between Fv-5 and Fv-6. Tibia, Td-6 longer than Td-1 but shorter than Td-2. Td-4 shorter than Td-2 and longer than Td-6. Td-3 shorter than Td-5. Tv-4 longer than Tv-6 and Tv-1. Tv-2 longer than Tv-5. Basitarsus, Bd-1 long, more than half the length of Bd-3. Bv-3 longer than Bv-1 but short. Bv-1 less than half the length of Bd-3. Pedipalp tarsus and post-tarsus completely fused. Tarsus without small spine on proximal end of dorso-inner lateral surface. Femur 5.3 mm long; tibia 7.1 mm long, 2.7 mm wide; basitarsus 3.4 mm long, 2.2 mm wide; tarsus 4.0 mm long. Legs. Second tarsomere of tarsi II and III with light transverse complete line on distal end. Tarsus IV is missing. Antenniform leg: 16.0 mm, femur; rest is missing. Leg II: 1 1 .0 mm, femur; 11.0 mm, tibia. Leg 111:12.0 mm, femur; 12.5 mm, tibia. Leg IV: 10.0 mm, femur; 1 1.0 mm (4. 8/0. 8/1. 9/3. 5), tibia; tarsus is missing. Female Genitalia (Barbados). As in Fig. 153. Natural History.— Nothing is known of the habits of this species. Distribution.— Barbados and St. Vincent. Figs. 78-89. -Pedipalps of male holotypes: Phrynus barbadensis (Pocock), 7 8-83; Phrynus gervaisii (Pocock), 84-89. Figs. 78, 85 = femur, dorsal view; 81, 88 = femur, ventral view; 79, 86 = tibia, dorsal view; 82, 89 = tibia, ventral view; 80, 87 = basitarsus and tarsus, inner lateral view; 83, 84 = tro- chanter, anterodorsal view. QUINTERO- THE AMBLYPYGID GENU S PHR YN US 147 Phrynus gervaisii (Pocock) Figs. 84-89, 113, 147, 148, 152, 167; Map 2 Phrynus palmatus Koch 1841 (nec Herbst, 1797):13-15, pi. 257. Fig. 601. Tarantula gervaisii Pocock 1894:285-6, pi. 7, Fig. 5. Male holotype from Magdaleine, Colombia, in the BMNH, examined. Neophrynus palmatus barbadensis: Kraepelin 1895:30-34 (in part). Tarantula palmata: Kraepelin 1899:242-244 (in part). Phrynus caracasanus Pereyaslawzewa 1901:117-304. Not Simon. Pereyaslawzewa described only em- bryo sections. Female at Museum Nat. d’Hist. Nat., Paris, examined, jar No. 37 labelled “ Phrynus caracasanus E. Simon TYPE, Caracas, Simon, 1899. Egg sac sent to Pereyaslawzewa.” ICZN, art. 9(5), labelling a specimen in a collection does not constitute publication. Phrynus gervaisii'- Mello-Leitao 1931:41. Diagnosis.-It is most closely related to Phrynus barbadensis. Phrynus gervaisii can be recognized by its poorly defined frontal area, its limits not distinct from rest of carapace, and by the dark variegated abdomen. A Trinidad specimen of P. gervaisii has a well- defined frontal area, distinct from the rest of carapace, but it is distinguished from P. barbadensis by its dark, variegated abdomen; P. barbadensis have a pale, yellowish abdo- men, with scarcely a trace of pattern. Description.— Male holotype. Carapace and pedipalps dark reddish-brown. Carapace with distinct flavous marginal spots on its posterior edges. Three flavous bands dorsal on Figs. 90-1 01. -Pedipalps of male holotypes: Phrynus whitei Gervais, 90-95; Phyrnus parvulus Po- cock, 96-101. Figs. 90, 97 = femur, dorsal view; 93, 100 = femur, ventral view; 91, 98 = tibia, dorsal view; 94, 101 = tibia, ventral view; 92, 99 = basitarsus and tarsus, inner lateral view; 95, 96 = trochanter, anterodorsal view. 148 THE JOURNAL OF ARACHNOLOGY leg IV, less clearly defined on the other ambulatory legs. Small flavous tubercles scattered on posterior half of carapace of dorsum of abdomen. Two short yellow lines beside lateral eyes. Abdomen variegated, some tergites with a flavous ring around the muscular impres- sions. Total length 17.3 mm. Carapace. With narrow, nearly straight anterior edge with small, even setiferous tuber- cles. Frontal process concealed from above. Poorly defined frontal area. Distance of median ocular tubercle from anterior edge half the length of tubercle (0.3 mm/0.6 mm). Carapace 6.9 mm long, 1 1.7 mm wide, 4.2 mm sulcus from anterior edge. Median ocular tubercle 0.7 mm wide. Lateral eyes 4.5 mm from each other, 1.2 mm from anterior edge, 1.3 mm from lateral edge. Chelicerae. Sparsely granular, anterodorsal surface of basal cheliceral segment with the external distal tubercle slightly enlarged. One well-developed tooth and a strong, uneven surface, sharp-edged ridge on external margin of basal cheliceral segment. Ridge appears to join the external tooth with the inner double-pointed proximal tooth (Fig. 1 13). Genital operculum. 3.0 mm long, 5.9 mm wide. Opisthogeminate organ as in Figs. 147 and 148. Abdomen 11.0 mm long. Trichobothria. As in Fig. 167. Ratios: sbc 0.44, bt 0.37. Pedipalps. Figs. 84-89. Robust in appearance. Trochanter with four spines, and a well-developed spine at center of anterior surface. Femur with numerous coarse granules on its dorsal surface, sparsely granulated ventrally. Femur Fd-2 longer than Fd-3. Fd-4 small, shorter than Fd-5 and Fd-6. Fv-1 and Fv-2 present a common base separate from Fv-3. Fv-4 one-third length of Fv-6. Fv-3 longer than Fv-6. Tibia, Td-6 longer than Td-1 but shorter than Td-2. Td-4 shorter than Td-2 and longer than Td-6. Td-3 shorter than Td-5. Tv-4 same size as Tv-6 but thinner. Tv-4 shorter than Tv-1. Basitarsus, Bd-1 short, about one-third the length of Bd-3. Bv-3 longer than Bv-1 which is reduced to an incon- spicuous tubercle. Bv-1 shorter than Bd-1. Pedipalp tarsus and post-tarsus completely fused. Tarsus without small spine on proximal end of dorso-inner lateral surface. Femur, Figs. 102-107. -Pedipalp of male lectotype Phrynus operculatus Pocock: 102, femur, dorsal view; 103, femur, ventral; 104, tibia, dorsal view; 105, basitarsus and tarsus, inner lateral view; 106, tibia, ventral view; 107, trochanter, anterodorsal view. QUINTERO-THE AMBLYPYGID GENUS PHRYNUS 149 6.9 mm long; tibia, 7.5 mm long, 2.9 mm wide; basitarsus, 3.8 mm long, 2.3 mm wide; tarsus, 4.1 mm long. Legs. Second tarsomere of tarsus 2 with light transverse complete line on distal end. Tarsi 3 and 4 have been lost. Antenniform leg: 17.4 mm, femur; 28.5 mm, tibia: 31.0 mm, tarsus. Leg II: 12.1 mm, femur; 17.8 mm, tibia. Leg III: 13.0 mm, femur; tibia lost. Leg IV: 10.2 mm, femur; 16.0 mm (7.0/1. 2/2. 3/5. 5), tibia; tarsus lost. Figs. 108-1 16. -Teeth on basal cheliceral segment, external view, right chelicerae: 108 P. operculatus, Pocock; 109, P. whitei Gervais; 110, P. parvulus Pocock; 111,P. barbadensis (Pocock); 112, P. santarensis (Pocock); 113, P. gervaisii (Pocock); 114, P. tessellatus (Pocock); 115, P. goesii Thorell; 116,P. longipes (Pocock). 150 THE JOURNAL OF ARACHNOLOGY Female Genitalia (Madden Forest Preserve, Canal Zone). As in Fig. 152. Natural History.— Epigean species found mainly in covered, secondary forests and occasionally associated with abandoned, dark places, moist objects near human dwellings. It has been collected from under boards in trash piles in Venezuela (Edo. Guarico. near Calaboso) and in the Canal Zone. Numerous collections of Phrynus gervaisii in Madden Forest Preserve, Canal Zone, found it commonly ocult during the day above the forest ground in the narrow space between the broad bases of the fronds and the trunk of the corozo palm, Scheelea zonensis Bailey. Adults and immatures have been found under the accumulated pile of fallen corozo fronds and fruits under the palm. It has also been found under the bark of large trees and under fallen, rotten logs near creeks and in the banks of rivers. Few individuals have been collected from the floor of the main chamber of Chili- brillo Cave in Panama, a cave where the top invertebrate predator is Paraphrynus laevi - frons (Pocock) (personal observations). About their reproductive biology it is known that they reproduce all year round and that females probably lay one egg case each year. Female carry egg cases with embryos, number varying from 9-24, mean of 15. Larger females generally carry a larger number of embryos. The exact gestation period is not know. Captured females that laid egg cases in captivity had young born in five to six months after being brought to the laboratory. I have observed their molting in capitivity which occurs during the night while hanging in an inverted position from any ledge. Shortly after ecdysis, the animal hides and is very pale (greenish), soft and humid and does not eat for two or three days. The exuvia remains hanging from the ledge. Nothing is known about their feeding preferences in nature. In captivity adults accept readily roaches and crickets and larval stages have been fed with termites. Distribution.-Costa Rica, Colombia, Ecuador, Venezuela, Guyana, Trinidad, and Panama. Phrynus whitei Gervais Figs. 90-95, 109, 143, 145, 146, 163; Map 1 Phrynus whitei Gervais 1842:19-22. Male holotype erroneously ticketed Burdwan, Bengal, India (Hardwicke’s collections) in the BMNH, examined. Gen. Thomas Hardwicke donated his collec- tions in 1835 to the BMNH, it is known his specimen labels include numerous erroneous locations. Pocock 1902a:50, 52-53, pi. 11, Figs. 1, la-c. Tarantula whitei : Pocock 1894:277, pi. 7, Figs. 4, 4a; Kraepelin 1899:243 (in part). Neophrynus whitei : Kraepelin 1895:28-29 (in part). Diagnosis.-The distal displacement of the dorsal spines on the- pedipalp tibia and the presence on the carapace of shiny or yellow patches beside the lateral ocular clusters are unique features of Phrynus whitei. It differs from Phrynus operculatus and P. parvulus in having Td-5 shorter than Td-3, and Td-2 and Td-6 about the same size. Description— Male holotype. Carapace and pedipalps dark reddish-brown. Carapace with a yellow rim, wider at posterior edges, and diffuse yellowish markings irradiating from sulcus. A very distinctive, shiny, silver-yellowish patch on the inner side of each lateral ocular cluster. Abdomen variegated with dark brown muscular impressions and a yellow band behind them. Femora of legs banded. Total length 14.0 mm. It does not appear to be a full-grown adult. Male specimens from La Ceiba, Honduras, reach 22.0 mm long. QUINTERO-THE AMBLYPYGID GENUS PHR YNUS 151 Carapace. Narrow, lightly emarginated anterior border, with very small, uneven size tubercles. Frontal process concealed from above, with sparsely arranged coarse granules. Distance of median ocular tubercle from anterior edge equal to length of tubercle (0.3 mm/0.3 mm). Carapace 5.8 mm long, 8.2 mm wide, 3.3 mm sulcus from anterior edge. Median ocular tubercle 0.7 mm wide. Lateral eyes 2.7 mm from each other, 1.0 mm from anterior edge, 0.7 mm from lateral edge. Chelicerae. One tooth on external margin of basal cheliceral segment (Fig. 109). Genital Operculum. 2.8 mm long, 4.2 mm wide. Opisthogeminate organ as in Figs. 143 and 145. Abdomen 9.0 mm long. Trichobothria. As in Fig. 163. Ratios: sbc 0.62, bt 0.30. Pedipalps. Figs. 90-95. Trochanter with four spines. Center of anterior surface without a spine. Femur, Fd-1 small, occult by Fd-2. Fd-2 longer than Fd-3. Fd-4 longer than Fd-5. Only five spines present. Fv-1 and Fv-2 with a common base separate from Fv-3. Fv-4 obsolete. Fv-3 same size as Fv-6. Tibia, Td-6 longer than Td-1. Td-2 same size as Td-6. Figs. 117-122. -Teeth on basal cheliceral segment, external view, right chelicerae: 117, P. pul- chripes (Pocock); 118, P. armasi new species; 119, P. asperatipes Wood; 120, P. marginemaculatus C. L. Koch; 121, P. levii, new species; 122, P. damonidaensis, new species. 152 THE JOURNAL OF ARACHNOLOGY Td-4 longer than Td-2 and Td-6. Td-3 longer than Td-5. Tv-4 about same size as Tv-6 and shorter than Tv-1. Tv-2 longer than Tv-5. Basitarsus, Bd-1 inconspicuous, a basal append- age to Bd-2. Bv-3 longer than Bv-1, both very reduced in size. Bv-1 longer than Bd-1. Pedipalp tarsus and post-tarsus completely fused. Tarsus without small spine on proximal end of dorso-inner lateral surface. Femur 5.2 mm long; tibia 5.9 mm long, 1.6 mm wide; basitarsus 2.7 mm long, 1 .6 mm wide; tarsus 2.9 mm long. Legs. Tarsi have been lost from all legs. Antenniform leg: 12.7 mm, femur; rest is missing. Leg II: 8.8 mm, femur; tibia broken. Leg III: 10.2 mm, femur; tibia broken. Leg IV: 8.4 mm, femur; 1 1.9 mm (5. 6/1. 0/2. 1/3. 2), femur; tarsus lost. Figs. 123-l.28.-P/zryttws pulchripes (Pocock): opisthogeminate organs, 123 = dorsal view, 124 = ventral view; 126 = female gonopods, dorsal view. P. armasi, new species: opisthogeminate organs, 127 = dorsal view, 125 = ventral view; 128 = female gonopods, dorsal view. QUINTERO-THE AMBLYPYGID GENUS PHRYNUS 153 Female Genitalia (Guanacaste, R. Santa Rosa, Costa Rica). As in Fig. 146. Variation.— The presence of a light transverse complete line on the distal end of the second tarsomere of all legs is variable and does not appear to be geographically related. The presence of the patch on the inner side of each lateral ocular cluster is a variable character, most specimens either have the shiny yellowish patch or a plain yellow patch; in others, this patch is almost absent or difficult to recognize. Natural History.— It has been collected from under old logs in a pine forest southwest of La Lima, Honduras, and Cueva de Las Pina Ramas, in Chiapas, Mexico. Distribution— Mexico: states of Jalisco. Veracruz and Chiapas. Guatemala, Honduras, El Salvador, Nicaragua, Costa Rica: Guanacaste and Puntarenas. Phrynus parvulus Pocock Figs. 96-101, 110, 156, 158, 160; Map 1 Phrynus parvulus Pocock 1902a:50-52, pi. 10, Figs. 7, 7a, b. Male holotype from the ruins of Tikal, Peten, Guatemala in the BMNH, examined. Tarantula marginemaculata yucatanensis Werner 1902. Male holotype from 4 Belize, Yucatan in Naturhistorisches Museum Wien (type No. 113), examined. Specimen fragmented. Diagnosis.— Phrynus parvulus is the only species of Phrynus with clubiform setae and one of the most conspicuously variegated species, particularly in the abdominal tergites and legs. It lacks the small spine on the dorso-inner lateral surface of the pedipalp tarsus. It has Td-4 longer than Td-2, although their difference in size is small, particularly in younger specimens. Males of P. parvulus can be distinguished from those of P. operculatus by their sharp-pointed setae, small genital operculum and by having the fourth abdominal sternite straight. Description.— Male holotype. Carapace and pedipalps dark reddish-brown, carapace ornamented with pale reddish patches on each side of the middle line and with four diffuse yellowish marginal spots. Dorsum of femora of legs conspicuously banded. Ter- gites yellow along the posterior border, and with a large crescentic yellow patch around the dark-brown muscular impressions. Total length 16.0 mm. Carapace. Clubiform setae. Narrow, slightly emarginated anterior border, with very small, even-sized tubercles. Frontal process concealed from above. Distance of median ocular tubercle from anterior edge nearly the length of tubercle (0.4 mm/0.5mm). Cara- pace 6.0 mm long, 8.5 mm wide, 3.9 mm sulcus from anterior edge. Median ocular tubercle 0.7 mm wide. Lateral eyes 2.7 mm from each other, 1.1 mm from anterior edge, 1.0 mm from lateral edge. Chelicerae. One tooth and a ridge on external margin of basal cheliceral segment. The ridge appears to connect the external tooth with the inner double-pointed proximal tooth (Fig. 110). Genital Operculum. Small, fourth abdominal sternite straight, 2.3 mm long, 3.8 mm wide. Opisthogeminate organ as in Figs. 156 and 158. Abdomen 10.2 mm long. Trichobothria. As in Fig. 160. Ratios: sbc 0.63, bt 0.40. Pedipalps. Figs. 96-101. Covered with clubiform setae. Surface of femur with fine close granulation. Trochanter with four spines. Center of anterior surface without a spine. Femur, Fd-1 small, occult by Fd-2. Fd-2 longer than Fd-3. Fd-4 a small tubercle. Fd-5 longer than Fd-6. Fv-1, Fv-2 and Fv-3 without a common base. Fv-4 a short tubercle. Fv-3 shorter than Fv-6. Tibia, Td-6 longer than Td-1 and more than half the length of 154 THE JOURNAL OF ARACHNOLOGY Td-2. Td-4 longer than Td-2 and Td-6. Td-2 distinctly longer than Td-6. Td-3 same size as Td-5. Tv-4 same size as Tv-1, shorter than Tv-6. Basitarsus, Bd-1 inconspicuous, a basal appendage to Bd-2. Bv-3 longer than Bv-1. Bv-1 longer than Bd-1. Pedipalp tarsus and post-tarsus completely fused. Tarsus without small spine on proximal end of dorso-inner lateral surface. Femur 5.7 mm long; tibia 6.2 mm long, 1.9 mm wide; basitarsus 2.8 mm long, 1.9 mm wide; tarsus 3.0 mm long. Figs. 129-134 .-Phrynus longipes (Pocock): opisthogeminate organs, 129 = dorsal view, 130 = ventral view; 132 = female gonopods, dorsal view. P. goesii Thorell: opisthogeminate organs, 133 = dorsal view, 131 = ventral view; 134 = female gonopods, dorsal view. QUINTERO- THE AMBLYPYGID GENUS PHRYNUS 155 Legs. Covered with blunt spines and clubiform setae. Second tarsomere of tarsi with light transverse complete line on distal end. Antenniform leg: 13.1 mm, femur; 21.3 mm, tibia; 22.4 mm, tarsus. Leg II: 9.5 mm, femur; 13.0 mm, tibia. Leg III: 10.1 mm, femur; 15.1 mm, tibia. Leg IV: 9.2 mm, femur; 14.3 mm (6.4/1. 5/2. 6/3. 8), tibia; 1.9 mm (0.8/0.3/0. 1/0.7), tarsus. Female Genitalia (Tikal, Guatemala). As in Fig. 159. Natural History. Found among ruins of Mayan civilization and in a cave in Trece Aguas, Guatemala. Distribution. -Known from the type locality in Tikal, from Uaxactum, El Peten, Trece Aguas, Guatemala and Belize (British Honduras). Phrynus operculatus Pocock Figs. 102-108, 154,155,157,162; Map 1 Phalangium palmatum Herbst 1797:82-92, pi. 4, Fig. 2. See discussion on its identification at the end of the description. Phrynus palmatus: Latreille 1804: 136 (citation only). Phrynus operculatus Pocock 1902a:50, 52, pi. 10, Figs. 8, 8a-c. Male and female syntypes from Guatemala in the BMNH, examined. Lectotype male designated. Diagnosis.— Phrynus operculatus is the only species of Phrynus having distinct sexual dimorphism in the size of the genital operculum. Males have a large genital operculum and the fourth abdominal sternite bent while females have a normal size genital operculum and the fourth abdominal sternite straight. The female gonopods are very characteristic, with broad-based sclerites ending each in a thin hook bent upwards and ventrally. Males of P. operculatus can be distinguished from those of P. pan’ulus by their darker, less variegated look, larger genital operculum, shape of fourth abdominal sternite, and their sharp, pointed setae and spines. Description.— Male lectotype. Carapace and pedipalps dark reddish-brown. There is no red upon the frontal area on the carapace and the legs are uniformly colored reddish- brown, without banding. It lacks the marginal spots on the carapace. Tergites dark look- ing, with a very diffuse variegated pattern of dark brown and reddish-yellow. Total length 13.0 mm. Carapace. Covered with sharp, pointed setae. Narrow, almost straight anterior border, with very small, uneven size tubercles. Frontal process concealed from above. Distance of median ocular tubercle from anterior edge equal to length of tubercle (0.4 mm/0.4 mm). Carapace 5.8 mm long, 9.0 mm wide, 3.3 mm sulcus from anterior edge. Median ocular tubercle 0.6 mm wide. Lateral eyes 2.7 mm from each other, 1.3 mm from anterior edge, 1.0 mm from lateral edge. Chelicerae. One tooth on external margin of basal cheliceral segment (Fig. 108). Genital Operculum. Large, fourth abdominal sternite bent. 4.0 mm long, 5.5 mm wide. Opisthogeminate organ as in Figs. 154 and 155. Abdomen 9.8 mm long. Trichobothria. As in Fig. 162. Ratios: sbc 0.52, bt 0.31. Pedipalps. Figs. 102-107. Covered with sharp, pointed setae. Trochanter with four spines. Center of anterior surface without a spine. Femur, Fd-1 small, occult by Fd-2. Fd-2 longer than Fd-3. Fd-4 a short tubercle. Fd-5 longer than Fd-6. Fv-1, Fv-2 and Fv-3 without a common base. Fv-4 obsolete. Fv-3 longer than Fv-6. Tibia, Td-6 longer than Td-1 and more than half the length of Td-2. Td-2 distinctly longer than Td-6. Td-4 longer 156 THE JOURNAL OF ARACHNOLOGY than Td-2 and Td-6. Td-3 same size as Td-5. Tv-4 shorter than Tv-1 and longer than Tv-6. Tv-2 longer than Tv-5. Basitarsus, Bd-1 inconspicuous, a basal appendage to Bd-2. Bv-3 shorter than Bv-1, both very reduced in size. Bv-1 longer than Bd-1. Pedipalp tarsus and post-tarsus completely fused. Tarsus without small spine on proximal end of dorso-inner lateral surface. Femur 5.2 mm long; tibia 5.9 mm long, 2.0 mm wide; basitarsus 2.9 mm long, 1.4 mm wide; tarsus 3.1 mm long. Legs. Covered with sharp, pointed spines and setae. Second tarsomere of all tarsi without light transverse complete line on distal end. Antenniform leg: 12.2 mm, femur; Figs. 135-140 -Phrynus damonidaensis new species: opisthogeminate organs, 135 - dorsal view, 136 = ventral view; 137 = female gonopods, dorsal view .P. levii, new species: opisthogeminate organs, 139 = dorsal view, 138 = ventral view; 140 = female gonopods, dorsal view. QUINTERO-THE AMBLYPYGID GENUS PHRYNUS 157 rest is missing. Leg II: 9.0 mm, femur; 12.4 mm, tibia. Leg III: 10.0 mm, femur; 13.5 mm, tibia. Leg IV: 8.4 mm, femur; 13.0 mm (5.6/0.9/2.3/4.2), tibia; 2.1 mm (0.9/0.3/0. 1/0.8), tarsus. Female Genitalia (Colima, Mexico). As in Fig. 157. Natural History.— It has been collected from under the bark of trees in Oaxaca and Jalisco, Mexico. In Oaxaca it has also been found under rocks in a pine-oak forest, and in a thorn forest under stones and under dead cactus. One female collected in Jalisco, Mexico, was carrying 40 praenymphs and 2 protonymphs, found under bark near a beach. Phrynus operculatus appears to be a strictly epigean species. Distribution. -USA: Texas, Big Bend area. Mexico: states of Nuevo Leon, Sinaloa, Nayarit, Jalisco, Guanajuato, Colima, Michoacan, Guerrero, Morelos, Oaxaca and Chiapas. Guatemala. On the Identity of Phalangium palmatum Herbst The identity of Phalangium palmatum Herbst, 1797, type species of the genus, has been the source of numerous contradictory opinions mainly because there was no collec- tion place indicated and the diagnostic features given by Herbst had to wait for a revision of the poorly characterized species to be recognized. The type specimen appears lost from the Zoologisches Museum der Humboldt, Berlin (Dr. M. Moritz, personal communica- tion). This was the institution where Herbst worked. I have identified Herbst’s palmatum as Phrynus operculatus Pocock 1902 on the following grounds: a. It has the typical “palmatus” look that Herbst so well described: “manus palporeim glaber, inflatus, quinque spinus palmatus” (p. 82). b. It has four spines on the dorsal border of the pedipalp femur. Herbst illustration on Fig. 2 of pi. 4 contradicts this point, showing five spines instead of four. I assume the drawing is not accurate in this aspect. Only three Phrynus species have four spines on the dorsal border of the pedipalp femur: P. whiitei with a peculiar silver design behind the eyes and a distal displacement of the spines dor sally on the pedipalp tibia, neither illus- trated or stated in Herbst’s work;P. parvulus, with a distinctive banding on the femora of the ambulatory legs, not stated nor illustrated by Herbst; and P. operculatus , that has Td-4 longer than Td-2 and the dorsal spines on the pedipalp tibia not distally displaced, as in Herbst’s illustration. c. Td-3 is slightly shorter or same size as Td-5 in Herbst’s illustration. Phrynus opercu- latus has Td-3 same size as Td-5. The only other species of Phrynus that comes close to the above description is Phrynus goesii : it has Td-3 slightly shorter than Td-5, Td-4 longer than Td-2, dorsal spines on pedipalp tibia not distally displaced, and no bandings on the femora of the ambulatory legs. But it has five spines on the pedipalp femur, and does not have the “palmatus” look, belonging to the group of the largest body size Phrynus species. The specimen illustrated by Herbst appears as a small Phrynus (no measurements were given). Because of the difficulty of explaining how a specimen of Phrynus operculatus (species that ranges from Texas, USA, to Guatemala) reached “the insectary of Baron von Block” in Kiel (now West Germany), where Herbst purchased specimens, and the contradictions with Herbst’s description indicated below, I had originally placed Phalangium palmatum as species incertae sedis. The contradictory details are: 158 THE JOURNAL OF ARACHNOLOGY 1. Distally to Td-5, Herbst mentioned there was only one small spine. No Phrynus species has only one spine distally to Td-5. P. operculatus has two spines, P. goesii has three. 2. Basitarsus of pedipalp with two spines dorsally and two ventrally in Phalangium palmatum. P. operculatus has a third, very small spine both dorsal and ventrally on the basitarsus; P goesii has three spines dorsally and three ventrally on the pedipalp basitarsus. The Law of Priority indicates that if two names are synonyms, then it is the older name that must be used as the valid name. Thus the name palmatum could not be legally rejected in favor of a junior synonym, operculatus. But it would have been an incorrect step against stability of nomenclature, as laid down in the preamble of the I.C.Z.N., to rename P. operculatus with an older name, one that has often been used incorrectly to identify several species of Phrynus, as it has been found on a large number of museum specimen labels, and that has a confused history of misidentifications. The description of operculatus has been used correctly for identifications since 1902 and the name describes very well the unique character of the species, male with large genital opercula. The “palmatum” character (digitated dorsal spines on the pedipalp tibia) is present in several species of Phrynus. The logic consequence of this argument is to request the Commission to use its plenary powers: (a) to supress the specific name palmatum Herbst, 1797, as published in the binomen Phalangium palmatum, for the purpose of the Law of Priority but not for those of the Law of Homonymy; and (b) to set aside all designations of type-species for the nominal genus Phrynus Lamarck, 1801, made prior to the Ruling now requested and, having done so, to designate Phrynus operculatus Pocock, 1902a, to be the type-species of that genus. Phrynus Incertae Sedis Phrynus pinarensis Franganillo 1930:48-49. Holotype from Sierra del Cuzco, Cordillera de los Organos, Pinar del Rio, Cuba probably unlabeled in the Academia de Ciencias de Cuba. Phrynus rangelensis Franganillo 1938:162, 1 male, 1 female. Syntype from Sierra de Rangel, Pinar del Rio and Baracoa, Oriente, Cuba, probably unlabeled in the Academia de Ciencias de Cuba. Phrynus viridescens Franganillo 1938:162-3. Holotype from Sierra de Rangel, Pinar del Rio, Cuba. I have been unable to examine the type material of Franganillo deposited in the Academia de Ciencias de Cuba because his specimens are unlabeled and marked only with numbers, but the catalogue is lost. Thus it is not possible to determine which specimen is the holotype or syntype. Franganillo descriptions are fragmentary at best and his species cannot be recognized. For Phrynus pinarensis, Franganillo described only the spination of the pedipalp tro- chanter and basitarsus and indicated that the subfrontal process was visible from above. The spination of the trochanter and the basitarsus does not allow species recognition. Of the four species described as present in Cuba ( Phrynus armasi, P. damonidaensis, P. marginemaculatus and P. levii), none has the subfrontal process visible from above. QUINTERO-THE AMBLYPYGID GENUS PHRYNUS 159 Phrynus rangelensis has the pedipalps black except the distal end of the basitarsus and tarsus which are red. The closest looking species from Cuba is P. marginemaculatus because the other species, P. armasi, P. damonidaensis, andP. levii are light colored. Like P. marginemaculatus , P. rangelensis has Td-4 shorter than Td-2 but differs in having Fv-3 longer than Fv-6 and because P. marginemaculatus reaches a maximum length of 18.0 mm while P. rangelensis syntype female is 23.0 mm long and the syntype male is 19.0 mm long. The description of Phrynus viridescens is absurd, giving in part a description of a spider Figs. 141-146.— Phrynus marginemaculatus C. L. Koch: opisthogeminate organs, 141 = dorsal view, 142 = ventral view; 144 = female gonopods, dorsal view. P. whitei Gervais: opisthogeminate organs, 145 = dorsal view, 143 = ventral view; 146 = female gonopods, dorsal view. 160 THE JOURNAL OF ARACHNOLOGY Figs. 147-15 l.-Phrynus gervaisii (Pocock): opisthogeminate organs, 147 = dorsal view, 148 - ventral view; 152 = female gonopods, dorsal view. P. tessellatus (Pocock): opisthogeminate organs, 151 = dorsal view, 149 = ventral view. P. santarensis (Pocock): 150 = female gonopods, dorsal view. P. barbadensis (Pocock): 153 = female gonopods, dorsal view. QUINTERO-THE AMBLYPYGID GENUS PHRYNUS 161 NATURAL HISTORY OF PHR YNUS Phrynus marginemaculatus C. L. Koch is the only species of Phrynus that has some information published on its life history, ecology and behavior (Muma 1967, Weygoldt 1969, 1970, 1972). In the present section, I have summarized the natural history of the genus Phrynus and added original information. Most of the body of information has been supplied by collecting labels which lamentably are too frequently incomplete. Because of my own field and laboratory work in Panama, I have been able to learn about the life history of Phrynus gervaisii (Pocock). Figs. 154-159 .-Phrynus operculatus Pocock: opisthogeminate organs, 154 = dorsal view, 155 = ventral view; 157 = female gonopods, dorsal view. P. parvulus Pocock: opisthogeminate organs, 158 = dorsal view, 156 = ventral view. P. asperatipes Wood: 159 = female gonopods, dorsal view. 162 THE JOURNAL OF ARACHNOLOGY The particular habitat requirements of Phrynus appear not to have diversified much among the different species. Phrynus species are usually encountered in secluded, humid, and cool habitats. In general, open, sunny and dry places are unfavorable habitats. Collec- tions of Phrynus asperatipes Wood from under rocks in a sand dune area (Baja California Sur) represent the most xeric conditions under which a species of Phrynus has been found. Within favorable habitats, individuals tend to aggregate and numerous specimens could be collected from a single cave. No information is available on the densities they could reach. In the humid forest of Madden Forest Preserve, Canal Zone, I have found from one to sixteen individuals (average five) of Phrynus gervaisii living on the corozo palm, Scheelea zonensis Bailey. Seldom are pairs or adults with young found occupying the same space between the trunk and a base of a frond of the palm. They appear to be solitary of habit and this has been cooperated in captivity where each individual prefers to occupy a different refuge inside a cage. Available information on Phrynus armasi, P. longipes and P. levii suggests that they live in caves and probably are troglophiles. Except for P. longipes that frequently occurs outside of caves, it is not known whether or not the other species can also occur outside of caves. All three are light-colored species, but with normal eye formation. The other two light-colored species, P. asperatipes and P. damonidaensis, appear to be epigean in habits. Both have been found under rocks and P. asperatipes also along creeks and in a 160 bfQ2! Sbf y b+ 0.40 S be 063 sc, 0.71 I fJ sc 1, 2.6-9 /*> Sc 1,2,6-10 165 bfQ/3 Sbf ✓ pt0.4 7 rSsbc 0.44 sc, 0.73 Sc 2-4,7-12 161 /ft 039 162 bf 0./0 ' sbf - bfOI3 Sbf Stf | OPERCULATUS > ptO.49 r Sc ,0.76 Sc 1,2,6-H be S be 0.52 Sc, 075 A sf 1,2,6-10 SC 1,2,6-H marginEma- CULATUS 166 I y ptOS5 167 bfOlO ■ Sbf Stf y bt 025 be 'SbcOAO - sc, Q65 bfO.I4. y pt 0.49 xl /be ll sb \ I be 0.44 sff Sbc 0.56 \y Sc, 0.70 li sc, 0.7 1 sc I.2A L SC, 070 Sf 2,6.7,9-11 !$ Sc 1.2.6,7,9-12 jjT sc h36i/ 7,9-11 A sc 1.2,6,7.9-12 1 (2.s/.,og sbf stf Sc, 0.7 Sf 1-3,6-10 /.V sc 1-3. 6t DAMONIDAENSIS 165 164 ' pt 0.51 y btojo . / pt 0.50 bf 0.13 sbf \ ✓ b DC UOt | ' stf se,Q7S s ”1*1 sc,07/ 12,6,8-11 /*\ Sc 1, 2. 6 f -1 2 Sc 3,4,6-10 )\ Sc 3,4, 6-H I6Q y pto.4 7 bf 0.1 7 V,/ be Sbf -4 169 bfOI4 Sbf Stf BARBADENSIS Spt 0J5! ,7 bt 0.40 I III Sbc0.45 Figs. 160-1 69. -Tibia of left leg IV, trichobothria present and ratios: 160, P. parvulus Pocock; 161, P. levii, new species; 162, Phrynus operculatus Pocock; 163, P. whitei Gervais; 164, P . damonidaensis, new species; 165, P. santarensis (Pocock); 166, P. margine macula tus C. L. Koch; 167, P. gervaisii (Pocock); 168, P. pulchripes (Pocock); 169, P. barbadensis (Pocock). QUINTERO-THE AMBLYPYGID GENUS PHRYNUS 163 palm oasis. The remaining ten species of Phrynus are dark colored. Of these, the following four appear to be strictly epigean in habits: P. tessellatus, P. marginemaculatus, P. san- tarensis, and P. operculatus. Phrynus marginemaculatus, P. santarensis, and P. gervaisii have been collected from occupied houses and particularly P. marginemaculatus is found commonly in Florida under abandoned trash left by man. Phrynus gervaisii less fre- quently is found in similar habitats, and occasionally on the floor of the Chilibrillo Cave in Panama. Phrynus santarensis from Santarem, Brazil, is the only species of the Phryni- dae that has been collected from a termite’s nest. The single record was given by Pocock in 1897. One species of Charinus and Paracharon caecus Hansen, both members of the Charontidae, have lost their eyes, are pale and live inside termite’s nests .Phrynus whitei and P. pulchripes have been collected from caves and also from under rotten logs in forests. Phrynus pulchripes has also been found under stones and under coconut husks on a beach. Phrynus species, like other amblypygids, are nocturnal in habits. During the night they come out from their hideouts in search of food and to mate. Nothing is known about their feeding habits in nature but from caged animals. They are raptorial predators and use their spiny pedipalps like spiny cages to capture and hold the prey, frequently still intact after capture, while they dig into them with their chelicerae and carry out a preoral liquefaction of the prey. For orientation and search for prey and water, they use their antenniform legs to carry sensory information, as whips during aggressive interactions, and during the prolonged courtship (Weygoldt 1969; personal observations on P. gervaisii). Although no published data is available on visual acuity, I have carried out some experi- ments covering the eyes (medial and lateral ocular groups) of P. gervaisii and found out that, when offered a prey (cricket or roach), the blindfolded animal can continue to capture prey without apparent difficulty. They frequently autotomise or fragment their antenniform legs without harm to the animal. If both antenniform legs are lost, they cannot capture prey, but such lost legs will be fully regenerated during the nest ecdysis. 170 •j' pt 0.45 y bt 0.40 bf 0./4_r jy t)C Sbf ~\ stf S be 0.57 I E-SC, 0.74 sf 1.2,6. 1?; I 7.9-14 jX SC /. 2.6,7. 9-/5 171 tr / bt 025 Sbf Stf . vjy be S be 0.54 ■ Sc. 0.72 Sf 4.5.6. |.f 1 11-17 /A SC 1.2.3,8-17 172 J pt 0.3 7 ,/bt 0.57 bfO.I7 be Sbf Sf 2, 4.6.9, 10 12.14-17 sc, o.7 a Sc 2.3,57 9-11. 13-I£ I 74 ARMASI ' pt 0.31 Jb bf 0.1 5 Sbf | \t SC'08° Sf 4-8,1 H6 \ji sc 1.5.8-17 Figs. 17 0-1 7 4. -Tibia of left leg IV, trichobothria present and ratios: 170, P. tessellatus (Pocock); 171, P. asperatipes Wood; 172, P. longipes (Pocock); 173, P. goesii Thorell; 174, P. armasi, new species. 164 THE JOURNAL OF ARACHNOLOGY Although adults of Phyrnus lack pulvilli on the tarsi of their ambulatory legs and thus cannot walk on smooth surfaces, the praenymphs do have pulvillus-like projections on the tarsi of their ambulatory legs. These are the stages that cling around the mother’s abdo- men after birth. These projections might serve at this stage to get a better grasp of their mother’s abdomen (Quintero 1975) and are lost when the protonymphs are formed, leaving their mother to start their free lives. For a description of the mating behavior of Phrynus marginemaculatus , see Weygoldt (1969). My observations on the mating of P. gervaisii found it takes place in a remarkably similar way. Several weeks or months after mating, Phrynus marginemaculatus lays a batch of eggs inside a brood sac. This is lentil-shaped and fits between the concave sides of the opistho- soma. The anterior end of the sac is held by the claw-like sclerites of the female gono- pods, under the genital operculum. The sac hardens in about 12 hours and becomes gray-brown and with two or three layers. Eggs average 1 .5 mm in diameter. The number of eggs varies according to the species and the size of the female. Muma(1967) found it varies from 17 to 36, with a mean of 24, for Phrynus marginemaculatus. It varies from 7 to 50, with a mean of 22, for P. damonidaensis. For P. gervaisii it varies from 9 to 24, with a mean of 15. During the development of the embryos the female is free to walk around and to capture prey. The development of the embryos in P. marginemaculatus takes from 91 to 105 days (Weygoldt 1970). The praenymphs, unpigmented, come out of the brood sac, an operation which lasts for several hours and cling to their mother’s abdomen. The first free-living stage, the protonymph, leaves immediately the body of the mother after molting. They are still light-green in color. ACKNOWLEDGMENTS I am most grateful to Dr. Herbert W. Levi for his assistance, encouragement and advice with the major aspects of this study and for his patience with its protractedness. Professor Edward O. Wilson served as major professor and offered support in many academic matters. I am sincerely grateful to him. Valuable advice was received from Dr. Michael H. Robinson, of the Smithsonian Research Tropical Institute, the editor and two reviewers of the Journal of Arachnology who critically read the manuscript. Much encouragement to carry this work to comple- tion was received from Dr. Ira Rubinoff, Director of the Smithsonian Tropical Research Institute; this institute provided air transportation for two of my visits to London to examine the BMNH collections. My sincere appreciation to Mr. Fred Wanless and Mr. Keight H. Hyatt of the British Museum (Natural History) for the many attentions received during two visits to that institution in 1975 and for the three loans of specimens from the collection under their care. Important collections were supplied by the following persons from collections under their care: Dr. H. W. Levi (Museum of Comparative Zoology), Drs. John A. L. Cooke and Norman Platnick (American Museum of Natural History), Dr. Paul Arnaud (California Academy of Sciences), Dr. Ralph B. Crabill (U.S. National Museum), Dr. Beatrice R. Vogel (Texas Memorial Museum), Dr. C. L. Hogue (Los Angeles County Museum), Dr. Joseph A. Beatty (Southern Illinois University), Dr. Stewart B. Peck (Carleton University, QUINTERO-THE AMBLYPYGID GENUS PHRYNUS 165 Ottawa), Dr. Robin E. Leech and Dr. Charles Dondale (Canada Department of Agricul- ture), Mr. Luis F. de Armas (Academia de Ciencias de Cuba), Dr. Carlos E. Valerio (Universidad de Costa Rica), Mr. Thomas H. Farr (Science Museum, The Institute of Jamaica), Dr. Anna T. Da Costa (Rio de Janeiro Museum), Dr. H. Fechter (Zool. Staats- sammlung, Munich), Dr. Jacqueline Heurtault (Paris Museum), Dr. L. Van der Hammen (Rijksmuseum, Leiden), Dr. Gisela Rack (Zoologisches Museum, Hamburg), Dr. M. Grass- hoff (Senckenberg Museum, Frankfurt), Dr. Ludwig Beck (Ruhr Universitat, Bochum), Dr. P. W. Hummelinck (Zoologische Laboratorium, Utrecht), Mr. T. Kronestedt (Natur- historiska Riksmuseet, Stockholm), and Mr. J. Kekenbosch (Institut Royal des Sciences Naturelles de Belgique). LITERATURE CITED Blanchard, E. 1852-1864. L’Organisation du regne animal. 2 Livre, Arachnida, pp. 168-201, pi. 10-11. Browne, P. 1789. The Civil and Natural History of Jamaica. London, Vol. 2: 419-420, pi. 41, Fig. 3. Butler, A. G. 1873. A monographic revision of the genus Phrynus. Ann. Mag. Nat. Hist., 12:117-125. Delle Cave, L., and M. Simonetta. 1975. Taxonomic notes on the Amblypygi (Arachnida, Chelicerata) from Ethiopia and Somalia. Monitore Zool. ital. (N.S.), suppl., 6:141-166. Fabricius, J. C. 1793. Etomologia Systematica, 2:407-438. Franganillo, B. P. 1930. Mas aracnidos nuevos de la isla de Cuba. Mem. Inst. Nac. Invest. Cient. Mus. Hist. Nat., La Habana, 1:45-99, 21 Figs. Franganillo, B. P. 1938. Aracnidos Cubanos estudiados desde 1930 hasta 1934. Mem. Soc. Cubana Hist. Nat., 8:145-168, 4 Figs. Gervais, M. P. 1842. Sur le genre Phrynus et Solpuga. Bull. soc. Philomatic, Paris, 5:19-22. Gervais, M. P. 1844. Histoire Naturelle des Insectes Apteres. Walkenaer III: 1. Herbst, J. F. 1797. Natursystem der ungefliigelten Instecten, 1 :65-86, pi. 3-5. Karsch, F. 1879. Ueber eine neue Eintheilung der Tarantuliden (Phrynidae aut.). Archiv. f. Natur., 45:189-197. Koch, C. L. 1841. Die Arachniden. Nurnberg, 8:1-131, pi. 253-288. Kraepelin, K. 1895. Revision der Tarantuliden Fabr. (Phryniden Latr.). Abh. Nat. ver. Hamburg, 13:1-53, 1 pi. Kraepelin, K. 1899. Skorpiones und Pedipalpi. Pedipalpi. Das Tierreich, 8. R. Friedlander und Sohn, Berlin. 265 pp. Lamarck, J. B. 1801. Systeme des Animaux sans vertebres. Paris. First ed. 432 pp. Linnaeus, C. 1758. Systema naturae. 10th ed. Holmiae. 821 pp. Lonnberg, E. 1897. Skorpioner och Pedipalper. I. Upsala Universitats Zoologiska Museum. Ent. Tidskr., 18:175-192. Mello-Leitao, C. 1931. Pedipalpos do Brasil e algumas notas sobre a Ordem. Arch. Mus. Nac. Rio Janeiro, 33:7-72, 3 pi. Mello-Leitao, C. 1946. Nuevos aracnidos sudamericanos de las colecciones del Museo de Historia Natural de Montevido. Comun. zool. mus. Hist. Nat., Montevideo, 2(35) : 1-1 0. Mullinex, C. L. 1975. Revision of Paraphrynus Moreno (Amblypygida: Phrynidae) for North American and the Antilles. Occ. Papers Calif. Acad. Sci., 116:1-80, 30 figs., 1 table. Mullinex, C. L. 1979. A new Paraphrynus from Yucatan (Amblypygida, Tarantulidae). J. Arachnol., 7:267-269. Muma, M. H. 1967. Scorpions, whip scorpions and wind scorpions of Florida. Arthropods of Florida, Florida Dept. Agric., 4:1-28. Pereyaslawzewa, S. 1901. Development embryonnaire des Phrynes. Annales. Sci. nat., Zoologie (8), 13:117-304. Pocock, R. I. 1893. Contributions to our knowledge of the Arthropod fauna of the West Indies. Parts 1 and 3. Scorpiones and Pedipalpi, with a supplementary note upon the freshwater Decapoda of St. Vincent. J. Linn. Soc., Zool., 24:404-409; 527-544, 2 pi. 166 THE JOURNAL OF ARACHNOLOGY Pocock, R. I. 1894. Notes on the Pedipalpi of the family Tarantulidae contained in the collection of the British Museum. Ann. Mag. Nat. Hist., 14(6): 27 3-298. Pocock, R. I. 1897. Reports upon the Scorpions and Pedipalpi. Ann. Mag. Nat. Hist., 19(6):357-368. Pocock, R. I. 1902a. Biologia Centrali-Americana: Arachnida, Scorpiones, Pedipalpi, and Solifugae. Taylor and Francis, London, 71 pp, pi. 1-12. Pocock, R. I. 1902b. A contribution to the systematics of the Pedipalpi. Ann. Mag. Nat. Hist., 9:157-165, Quintero, D. 1975. Scanning electron miscrope observations on the tarsi of the legs of the amblypy- gids (Arachnida, Amblypygi). Proc. 6th Int. Arachn. Congr., 1974: 161-163, 2 pi. Quintero, D. 1976. Trichodamon Mello-Leitao and the Damonidae, new Family status (Amblypygi: Arachnida). Bull. Birt. Arachnol. Soc., 3(8):222-227. Quintero, D. 1979. Sobre Paraphrynus emaciatus Mullinex,/>. leptus Mullinex (Amblypygi: Phrynidae) y el dimorfismo sexual en los amblypygidos. Cuaderno de Ciencias No. 3:15-24. Ed. Universitaria, Panama. Quintero, D. 1980. Systematics and Evolution of Acanthophrynus Kraepelin (Amblypygi, Phrynidae). Proc. 8th Int. Arachn. Congr. :341-347. Simon, E. 1892. Arachnides des lies Philippines. Appendice. Remarques sur la classification des Pedipalpes de la Famille des Tarantulidae. Ann. Soc. ent. France. 61 :43-52. Simon, E. 1897. On the Spiders of the Island of St. Vincent. Part 3. Proc. Zool. Soc. London :860-890. Thorell, T. 1889. Arachnidi Arthrogastri Birmani. Annali Mus. Civico Storia Nat. Genova, 7(2):521-729, pi. 5. Werner, Fr. 1902. Die Scorpione, Pedipalpen und Solifugen in der Zoologisch-vergleichind- anatomischen Sammlung der Universitat Wien. “Verhandlungen” der K.K. zool-bot. gesellschaft in Wien, 595-608. Weygoldt, P. 1969. Beobachtungen zur Fortpflanzungsbiologie und zum Verhalten der Geisselspinne Tarantula marginemaculata C. L. Koch (Chelicerata, Amblypygi). Z. Morph. Tiere, 64:338-360. Weygoldt, P. 1970. Lebenszyklus und postembryonale Entwicklung der Geisselspinne Tarantula mar- ginemaculata C. L. Koch (Chelicerata, Amblypygi) im Laboratorium. Z. Morph. Tiere, 67:58-85. Weygoldt, P. 1972. Morphologisch-histologische Untersuchungen an den Geschlechtsorganen der Am- blypygi unter besonderer Berucksichtigung von Tarantula marginemaculata C. L. Koch (Arac- hnida). Z. Morph. Tiere, 73:209-247. Weygoldt, P. 1975. Untersuchungen zur Embryologie und Morphologie der Geisselspinne Tarantula marginemaculata C. L. Koch (Arachnida, Amblypygi, Tarantulidae). Zoomorphologie, 82:137-199. Wood, H. C. 1863a. Description of new species of North American pedipalpi. Proc. Acad. Nat. Sci., Philadelphia, 107-112. Wood, H. C. 1863b. On the pedipalpi of North America. J. Acad. Nat. Sci., Philadelphia, 5:357-376. Manuscript received July 1979, revised February 1980. Millidge, A. F. 1981. The erigonine spiders of North America. Part 3. The genus Scotinotylus Simon (Araneae, Linyphiidae). J. Arachnol., 9:167-213. THE ERIGONINE SPIDERS OF NORTH AMERICA. PART 3. THE GENUS SCOTINOTYLUS SIMON (ARANEAE: LINYPHIIDAE) A. F. Millidge Little Farthing, Upper Westhill Road, Lyme Regis, Dorset DT7 3ER, England ABSTRACT A revision has been carried out of the North American members of the genus Scotinotylus Simon. The generic names Caledonia Cambridge, Cervinargus Vogelsanger, Cheraira Chamberlin, Cochlem- bolus Crosby, Coreorgonal Crosby and Bishop and Yukon Chamberlin and Ivie are synonymized with Scotinotylus. Araeoncus patellatus Emerton, Ceratinopsis eutypa Chamberlin, Disembolus apache Chamberlin and “Erigone” bodenburgi Chamberlin and Ivie have been transferred to Scotinotylus, while Cochlembolus sacerdotalis Crosby and Bishop and Cochlembolus provo Chamberlin have been excluded from the genus. Scylaceus divisus Chamberlin is a synonym of Scotinotylus vernalis (Emerton), Cheraira willapa Chamberlin is a synonym of Scotinotylus monoceros (Simon), and Spirembolus chera Chamberlin and Ivie and Cheraira salmonis Chamberlin are synonyms of Scotinotylus sanctus (Crosby). The revised genus Scotinotylus has been defined chiefly on the struc- tures of the male palpal organs and the female epigyna; synapomorphic genitalic characters have been identified. The genus contains 34 species in North America, including the following 17 new taxa: Scotinotylus ambiguus, S. bicavatus, S. bipoculatus, S. boreus, S. crinitus, S. dubiosus, S. exsectoides, S. gracilis, S. humilis, S. magnificus, S. montanus, S. petulcus, S. pollucis, S. regalis, S. sacratus, S. sagittatus and S. sintalutus. The genus is subdivided into three species groups, the antennatus, kenus and monoceros groups. Members of the genus are distributed throughout the cooler latitudes of the northern hemisphere, but the majority of the species appear to be endemic to North America. Decrip- tions, diagnoses and distribution maps are given for each species. INTRODUCTION The genus Scotinotylus was erected by Simon (1884) for the two European species Erigone antennata Cambridge and Erigone alpigena L. Koch. In the present revision of the North American species of this genus, the generic names Caledonia Cambridge 1894, Cochlembolus Crosby 1929, Coreorgonal Bishop and Crosby 1935, Cervinargus Vogel- sanger 1944, Yukon Chamberlin and Ivie 1947 and Cheraira Chamberlin 1948 are consi- dered to be junior synonyms of Scotinotylus ; the reasoning in support of this hypothesis is given later in this paper. Most of the material examined in this revision was loaned by the American Museum of Natural History, New York (AMNH), but a number of specimens were also supplied from the Museum of Comparative Zoology, Harvard University (MCZ) and the Canadian National Collection, Ottawa (CNC). 168 THE JOURNAL OF ARACHNOLOGY GENUS SCOTINOTYLUS SIMON Scotinotylus Simon 1884:502 (type species Erigone antennata Cambridge: “first species rule”) Scotynotylus : Roewer 1942:686 (this is not Simon’s original spelling) Caledonia Cambridge 1894:23. NEW SYNONYMY. (Type species C. evansi Cambr. by monotypy) Cochlembolus Crosby 1929:79. NEW SYNONYMY. (Type species Dismodicus alpinus Banks by original designation) Coreorgonal Bishop and Crosby 1935:217. NEW SYNONYMY. (Type species Delorrhipis bicornis Simon by original designation) Cervinargus Vogelsanger 1944:175. NEW SYNONYMY. (Type species C. prominens Vogelsanger ( =Tiso (?) clavatus Schenkel) by original designation) Yukon Chamberlin and Ivie 1947:52. NEW SYNONYMY. (Type species Y. majesticum Chamb. and Ivie by original designation) Cheraira Chamberlin 1948:518. NEW SYNONYMY. (Type species C. kena Chamb. by original desig- nation) The members of this genus are small spiders with a total length of 1. 2-3.0 mm. The female carapace is slightly elevated behind the eyes (Fig. 24) but is otherwise unmodified. The male carapace exhibits a diversity of forms, and in some species exaggerated lobes are present (e.g. Figs. 53, 86, 145). Those species which have a definite dorsal lobe usually have holes and sulci behind the lateral eyes, and the lobe in most cases does not carry the posterior median eyes (e.g. Fig. 39). In the type species, however, the rather shallow lobe carries the posterior median eyes, and lateral holes and sulci are absent (Fig. 17). All the species have files on the lateral margins of the chelicerae in both sexes. The abdomen is without scuta and is more or less unicolorous; there are clear striations on the epigastric plates in some species, particularly in the males. The legs in most species are relatively short and stout, with a value for tibia I 1/d (female) of 4.5-6; the larger species, however, tend to have somewhat thinner legs. In all species except one the tibial spines are 2221 in the female, usually reduced in number in the male; in S. formicarius (Dondale and Redner) the spines are 1111 in both sexes. The males of many of the species have short curved hairs dorsally on tibiae I and metatarsi I, as in Spirembolus Chamberlin (Millidge 1980). Metatarsi I-III have a dorsal trichobothrium, which is absent on metatarsus IV; the value of Tml lies in the range 0.35-0.70, but in the majority of species it is 0.35-0.55. In most of the species the male palpal patella is fairly short and unmodified, but in S. monoceros (Simon), S. bicornis Emerton and S. petulcus, new species, it is longer and swollen distally (Fig. 155). The male palpal tibia frequently bears one or more stout (thickened) spines dorsally. The tibial apophysis is usually fairly short, and frequently has a small tooth distally (e.g. Fig. 93), but in S. monoceros, S. bicornis and S. petulcus the apophysis is much longer and terminates in a hook (Fig. 155); there is a terminal hook also in the type species and in S. eutypus (Chamberlin), but in these species the apophysis is much shorter (Fig. 18). The female palpal tibia has 2 or 3 trichobothria dorsally; the male palpal tibia has 2. The characters given above are rather similar to those of the genus Spirembolus , and do not serve to distinguish Scotinotylus from related genera. In common with the majority of erigonine genera, therefore, a more exact definition of the genus must be based on the structure of the genitalia. The paracymbium of the male palp has a simple horseshoe shape in most species (e.g. Fig. 25), but is slightly more complex in a few species. In most cases the tegulum has an anterior projection which is distally whitish and somewhat membraneous (e.g. T, Figs. 2, 9). The embolic division (ED) is composed of a spiral embolus of two or more turns MILLIDGE-GENUS SCOTINOTYLUS 169 arising from a screw-like or tightly coiled tailpiece (e.g. Figs. 19, 83, 1 10). In four species (S. alienus (Kulczynski), the European species S. evansi (Cambridge), S. protervus (L. Koch) and S. kenus (Chamberlin)) there is a pointed apophysis arising from (or near) the first coil of the embolus (Figs. 45, 68, 94), and in S. regalis , new species, there is a lamellar apophysis arising from the base of the embolus (L, Fig. 90). The distal end of the embolus is normally simple, but in S. protervus the embolus terminates in a small loop (Figs. 67, 68). The suprategular apophysis (SA) is composed of (i) a weakly sclerotized tusk-like part of fairly constant form, which runs ventrad from the suprategulum (TK, Figs. 2, 9, 11); and (ii) a translucent membraneous part of variable form which arises from the mesal side of the “tusk” (M, Figs. 2, 9, 1 1). The membraneous stalk (S, Figs. 2, 9), which carries the seminal duct to the ED, is virtually a continuation of the membraneous part of the SA. The distal portion of the membraneous part has various forms. It may comprise a fairly simple sheet which lies across the anterior of the palp above the projecting tegulum (Figs. 1-6, 12-15), or it may be a lamella which is folded irregularly towards the ectal side (Figs. 8-11, 92, 108, 156). The range of variation of the SA within Scotinotylus (as here defined) is not regarded as excessive; some degree of variation within erigonine genera is Figs. 1-6. -Male palpal organs, meso-ventral view, with embolic division removed to show suprategular apophysis: 1, S. antennatus ; 2, S. evansi ; 3, S. alpinus’, 4, S. majesticus ; 5 ,S. protervus ; 6, S. vernalis. Abbreviations: M, membraneous part of suprategular apophysis; S, stalk; T, tegulum; TK, tusk -like part of suprategular apophysis (Scale lines 0.1 mm). 170 THE JOURNAL OF ARACHNOLOGY not unusual, e.g. in Mecopisthes Simon (Millidge 1977b), Diplocephalus Bertkau (Merrett 1963, Millidge 1977a) and Walckenaeria Blackwall (Merrett 1963). The form of the ED in Scotinotylus is generally similar to that in Spirembolus , but the tailpiece shows small differences; the form of the SA is quite distinct from that in Spirembolus. The combination of the form of the ED with the form of the SA can be regarded as a derived palpal character unique to the species of Scotinotylus , and this synapomorphy supports the hypothesis that the genus is monophyletic. The form of the epigynum is basically similar in all the species of the genus. Posteriorly there is outlined a roughly trapezoidal area, the “plate”; this narrows anteriorly to a median septum, which is clear in some species (e.g. Figs. 22, 95, 151) but less clear in others (e.g. Fig. 27). The markings visible on the epigynum, including the “plate”, are the positions of the more heavily sclerotized parts and of the internal apodemal structures which carry the spermathecae and ducts. There is a hollow on either side of the septum, anterior to the “plate”, and in these two cavities are situated the openings to the spermathecal ducts. The hollows are well developed and conspicuous in some species (e.g. Fig. 72) but rather obscure in others (e.g. Figs. 22, 27). In some species there is a median cusp-shaped knob (e.g. Fig. 55) or a linguiform process arising from the anterior margin of the epigynum; the “tongue” is well developed and conspicuous in a few species, e.g. S. majesticus Chamberlin and Ivie (Fig. 85) and S. patellatus (Emerton) (Fig. 61), but much shorter in others (e.g. Figs. 27, 31). Some species (those previously placed in Cheraira) have a cusp-shaped pocket on either side of the epigynum anterior to the “plate”; these cusps may be moderately large (Fig. 125) or very small (Fig. 127). It is perhaps worth noting that the median knob-like process mentioned above often carries within it a cusp-shaped hollow (Fig. 72). The internal genitalia of all the species have a similar basic plan. The spermathecal duct arises on the mesal side of the spermatheca, and for most of its course to the external opening it lies on the dorsal side of the spermatheca; the course followed by the duct may be relatively short and simple (Fig. 133) or longer and more sinuous (Figs. 73, 87). The external openings of the ducts are similarly placed to those of Spirembolus , and indeed the female of Scotinotylus monoceros has sometimes been mistaken for Spirembolus mundus Chamberlin (Millidge 1980). The arrangement of the ducts in Scotinotylus is however different from that in Spirembolus , where the duct follows a spiral course around the spermatheca. The total structure of the female genitalia, namely the form of the external epigynum with the duct openings in the hollows on either side of the median septum, coupled with the general positioning of the ducts, can be regarded as a derived epigynal character which is common to all species of the genus. This synapomorphy offers additional support to the hypothesis that the genus is a monophyletic group. Synonymy.— Previous to this paper it has already been suggested that Cochlembolus may be a junior synonym of Caledonia (Holm 1950), that Cheraira may be a synonym of Caledonia (Bragg and Leech 1972), and that both Cochlembolus and Caledonia are prob- ably junior synonyms of Scotinotylus (Millidge 1977a). In the present paper, Caledonia, Cochlembolus, Yukon, Cheraira and Coreorgonal are regarded as junior synonyms of Scotinotylus; Cervinargus, already a junior synonym of Cochlembolus (Thaler 1970), likewise falls into synonymy with Scotinotylus. All the species previously placed in these genera have basically similar male and female genitalia. Both sexes of Araeoncus patel- latus Emerton have the genitalia of this same form, and this species is also transferred to Scotinotylus. The three species Ceratinopsis eutypa Chamberlin, Disembolus apache MILLIDGE-GENUS SCOTINOTYLUS 171 Chamberlin and “Erigone” bodenburgi Chamberlin and Ivie have been transferred to Scotinotylus on the basis of their epigyna. The figure of the epigynum of Scironis autor Chamberlin (1948) indicates that this species probably belongs in Scotinotylus , but it has not been possible to include it in this study since the unique specimen on which the species was based cannot be located. The palpal organs of Caledonia evansi (the type of Caledonia ) (Figs. 44, 45) are very similar to those of Scotinotylus antennatus and its sibling S. eutypus (Figs. 18, 19), differing only in the shorter and stouter embolus in Caledonia and in a small difference in the membraneous part of the SA (Figs. 2, 13 cf. Figs. 1, 12). The palpal organs of Cochlembolus sacer Crosby (Fig. 26) and C. pallidus Emerton are very close indeed to those of S. antennatus , with the embolus long and thin distally and with the SA mem- brane very similar. The palpal organs of Cochlembolus alpinus (Banks) (the type of Cochlembolus) (Figs. 3, 51, 52), C vernalis (Emerton) (Figs. 6, 74) and Araeoncus patellatus (Figs. 58, 59) are closely similar to those of Caledonia evansi. Apart from a minor difference in the form of the tailpiece, the palpal organs of Yukon majesticum T 14 T if, 13 Figs. 7-15. -Male palpal organs, with embolic divisions removed to show suprategular apophysis: 7, S. formicarius, meso-ventral; 8, S. sanctus, meso-ventral; 9, S. monoceros, meso-ventral; 10, S. kenus, meso-ventral; 11, S. kenus, anterior view; 12, S. antennatus, anterior view; 13, S. evansi, anterior view; 14, S. majesticus, anterior view; 15, S. formicarius, anterior view. Abbreviations: M, membraneous part of suprategular apophysis; S, stalk; T, tegulum; TK, tusk-like part of suprategular apophysis (Scale lines 0.1 mm). 172 THE JOURNAL OF ARACHNOLOGY Chamberlin and Ivie (Figs. 4, 14, 82, 83) are also similar to those of Caledonia evansi. The Cheraira species (which included C. salmonis Chamberlin, a junior synonym of Coch- lembolus sanctus Crosby: see description of latter species later in this paper) have the ED of the same general form as that of Scotinotylus antennatus and Caledonia evansi , but with the SA membrane differently shaped. The type species of Cheraira (C. kena Cham- berlin) shares with Caledonia the character of a sclerotized tooth present on the basal coil of the embolus. Cochlembolus formicarius Dondale and Redner has the epigynum similar to those of the Cheraira species (presence of cusps) but the SA membrane closer to Caledonia than to Cheraira. Figs. 16-24.-16, 5. eutypus, male carapace, lateral; 17, 5. antennatus, male carapace lateral; 18, 5. eutypus, male palp, ectal; 19, 5. eutypus, male palp, mesal; 20,5. eutypus, epigynum; 21,5. eutypus, male palpal tibia, dorsal; 22, 5. antennatus, epigynum; 23, 5. antennatus, internal genitalia, female, ventral; 24, 5. eutypus, female carapace, lateral. Abbreviations: E, embolus; M, membraneous part of suprategular apophysis. (Scale lines 0.1 mm). MI LLIDGE- GENU S SCOTINOTYLUS 173 The Coreorgonal species have the ED of the same spiral form, but with the tailpiece slightly more complex distally (Figs. 157, 158); the SA is of the same basic form (Fig. 9), with the SA membrane fairly similar to that of Cheraira. The Coreorgonal males show greater differences from the other species in the following respects, however: (i) the tibial apophysis is much longer; (ii) the palpal patella is longer and swollen distally; (iii) there is a white excrescence between palpal patella and tibia, rather similar to that present between femur and patella in some species of Spirembolus ; and (iv) the form of the male carapace is different, with a lobe or stalk arising from the clypeal region. It is arguable whether these differences would justify the retention of Coreorgonal as a separate genus. The form of the tibial apophysis can certainly vary a good deal within a genus, while the lengthening of the palpal patella occurs in isolated species in several other genera (e.g. Erigone Audouin, Diplocephalus, Spirembolus). The excrescence between palpal patella and tibia is not necessarily of generic significance, since in the genus Spirembolus a somewhat similar excrescence is present in only a few otherwise typical species. The form of the male carapace can certainly show considerable variations within a genus, but it must be admitted that the form exhibited by Coreorgonal is very unusual. As mentioned earlier in this paper, the female genitalia of all the species under consi- deration are basically similar. Comparison of Fig. 47 ( Caledonia ) with Fig. 85 {Yukon) shows a clear similarity in form, the main difference being that the “tongue”, which is vestigial in Caledonia, is highly developed in Yukon. The epigynum of Araeoncus patel- latus, with its fairly long narrow “tongue” (Fig. 61), is in other respects close to those of Caledonia or Cochlembolus sacer (Fig. 27). It seems probable that the lengthening or shortening of the “tongue,” or its complete disappearance as in Cheraira and Coreorgonal , is of secondary importance. The presence on the epigyna of the Cheraira species of “cusps,” which in some species are very small, is probably also of secondary importance. The wavy outline of the internal apodemal structure which gives a characteristic appearance to the epigynum of Cochlembolus sacer (Fig. 27) is also present in two Cheraira species (Figs. 128, 130). Apart from the absence of “cusps,” the Coreorgonal species have the genitalia very similar to those of the Cheraira species. The Coreorgonal epigynum (Fig. 151) is also generally similar to that of the European species Scotinotylus alpigena (L. Koch) (Fig. 50), apart from the presence of the vestigial “tongue” in the latter species. The internal genitalia of all the species, including the Coreorgonal species, have the same basic pattern. On the basis of these data and considerations, it seems best to recognize the close relationships which clearly exist between the species placed in Scotinotylus, Caledonia, Cochlembolus, Yukon, Cheraira and Coreorgonal by uniting them into one genus: Scotinotylus Simon 1884 has priority. It must be admitted that the inclusion of the Coreorgonal species in this enlarged genus may be open to some question, but on balance, taking particularly into account the structures of the male and female genitalia, it seems a reasonable hypothesis to regard the three closely related species concerned as a slightly aberrant branch of Scotinotylus. Some arachnologists might prefer to retain certain of the smaller genera, in particular Scotinotylus, Cheraira and Coreorgonal. It is clear from the data presented, however, that the distinctions between these genera would be somewhat diffuse and ambiguous, and the genera would have to be defined and differentiated on the basis of small differences in the form of the SA membrane and on small differences in the female genitalia. If such minor variations from the type were to be regarded as generically significant in the present case, then consistency might demand the fragmentation of many commonly used erigonine genera into smaller genera, which I would regard as a distinctly retrograde step. 174 THE JOURNAL OF ARACHNOLOGY Species and species groups.— The genus Scotinotylus as defined in this paper contains 34 species in North America. It is helpful, for taxonomic purposes, to split this rather large genus into smaller groupings, which in the present state of knowledge are preferably designated as species groups rather than as sub-genera. The species groups are defined as follows: 1. The antennatus group comprises the species with the following characters: the SA of the male palp is distally a flattish membrane as shown in Figs. 1, 2, 12, 13; the epigynum has a cusp-like knob on the anterior margine (e.g. Fig. 20) or a tongue-like process (short or long) which arises from the anterior part of the epigynum (e.g. Figs. 31, 61) (exception S. ambiguus)', the female palpal tibia has 2 trichobothria dorsally; the species occur widely over N. America. In the absence of the male, the placing of S. ambiguus in this group is provisional. 2. The kenus group comprises the species with the following characters: the SA of the male palp is of the form shown in Figs. 8, 10, 11, 98, 108 (exception S. formicarius)’, the epigynum has a “cusp” on either side anteriorly; the female palpal tibia has 3 tricho- bothria dorsally (one may be very small or occasionally absent); the species are limited to the western half of N. America. 3. The monoceros group comprises the species with the following characters: the SA of the male palp is of the form shown in Fig. 9; the palpal patella is long and swollen distally (Fig. 155); the male carapace has a lobe or stalk arising from the clypeus (Figs. 145, 146, 147, 148); the epigynum has neither a tongue-like process nor “cusps” (Fig. 151); the female palpal tibia has 3 trichobothria; the species are limited to the western side of N. America. The species dealt with in this paper are as follows: antennatus species group Scotinotylus eutypus (Chamberlin) S. sacer (Crosby) S. pallidus (Emerton) S. sacratus , new species S. alienus (Kulczynski) S. alpinus (Banks) S. patellatus (Emerton) S. gracilis, new species S. protervus (L. Koch) S. vernalis (Emerton) S. exsectoides , new species S. majesticus (Chamberlin and Ivie) S. magni ficus, new species S. regalis, new species S. ambiguus , new species kenus species group S. kenus (Chamberlin) S. castoris (Chamberlin) S. pollucis, new species S. sanctus (Crosby) S. crinitus, new species S. montanus, new species S. humilis, new species MILLIDGE-GENUS SCOTINOTYLUS 175 S. bicavatus , new species S. bipoculatus , new species S. boreus, new species S. sintalutus, new species S. dubiosus, new species S. apache (Chamberlin) S. formicarius (Dondale and Redner) S. bodenburgi (Chamberlin and Ivie) S. sagittatus, new species monoceros species group S monoceros (Simon) S. bicornis (Emerton) S. petulcus , new species Misplaced species. -The following two species do not belong in Scotinotylus : Cochlembolus sacerdotalis Crosby and Bishop 1933:167. This species has been transferred to Disembolus Chamberlin and Ivie, and will be dealt with in Part 4 of this series of papers. Cochlembolus provo Chamberlin 1948:522. Examination of the type female (AMNH) shows clearly that this species is not a member of the genus Scotinotylus. Keys to species.— The genus contains several groups of closely related species which may exhibit only small structural differences. Partial keys to the species are presented in Tables 1 and 2; in all cases, the species descriptions and diagnoses should be referred to before a final identification is made. Distribution and Natural History.— The genus Scotinotylus (as defined in this paper) is widely distributed throughout the cooler regions of the northern hemisphere, but the majority of the known species appear to be endemic to North America. It is possible that North America represents the area of origin of the genus, and that those few species which are to be found outside North America have resulted from migrations from this focal area. The type species S. antennatus is known only from the European Alps and the Carpathians (Tatra Mountains), while its sibling species S. eutypus is known only from north-western America. Neither species has been found in such well searched areas as Greenland, Iceland, Scotland and Scandinavia. It seems not unlikely, therefore, that the dispersion route for this species pair has been from western North America via Asia to Europe. The European species S. clavatus (Schenkel), which is closely similar to S. sacer (Thaler 1970), is known from the European Alps only, while S. sacer is fairly widespread in north-western America and is also known from west Greenland (Holm 1967), but appears to be absent from Iceland and northern Eruope. This suggests that the dispersion to Europe in this case also may have been via Asia. The distribution of the species pair S. evansi/S. alienus is however circumpolar, though there are as yet no records from eastern Canada; S. evansi has also reached the European Alps, where it is apparently uncommon (Thaler 1970). Members of the genus have been collected from many parts of North America. The species seem on the whole to prefer the cooler climates of higher latitudes and/or altitudes, and there are no records from the south-eastern states of U.S.A. or from Mexico. Little has been recorded on the natural history of most of the species. Some sparse information indicates that they have the normal erigonine habits, and live at 176 THE JOURNAL OF ARACHNOLOGY Table 1. -Partial key to Scotinotylus species: males . 1. Carapace elevated into a small lobe, with 2 short curved spines projecting from the ocular area (Fig. 16); there are no holes or sulci behind the lateral eyes S. eutypus 2. Carapace lacking holes and sulci, but elevated and projecting anteriorly, and clothed with numerous hairs (Figs. 46, 60, 71) a. palpal tibia with 3 fairly stout spines S. patellatus b. palpal tibia lacking stout spines S. protervus ( S . alienus will also be keyed here) 3. Carapace with lobe or stalk arising from the clypeus, in addition to any other lobes a. carapace as in Fig. 145 S. bicornis b. carapace as in Fig. 146 S. petulcus c. carapace as in Figs. 147, 148 S. monoceros 4. Carapace with distinct dorsal lobe, which has sulci and holes behind the lateral eyes (e.g. Fig. 28); an additional lobe may also be present a. palpal tibia lacking stout spines i. one large and one smaller lobes present (Fig. 86); palp Figs. 82, 83, 84 S. majesticus ii. one large lobe and one tiny lobe present (Fig. 88) S. magnificus iii. a single large lobe present (Figs. 89, 99) S. regalis, S. pollucis (separate by palps) iv. a shallow lobe present (Fig. 76); palpal tibia Fig. 78 S. vernalis b. palpal tibia with one stout spine (e.g. Fig. 92) i. tibial apophysis fairly long and pointed (Fig. 142) S. formicarius ii. tibia with small black tooth on anterio-ectal margin (e.g. Figs. 93, 112) S. kenus, S. sanctus, S. crinitus , S. montanus, S. humilis (see species descrip- tions) c. palpal tibia with 2-3 stout spines (e.g. Fig. 25) i. clear epigastric striae present S. pallidus ii. epigastric striae absent or very indistinct S. alpinus, S. sacer, S. sacratus (separate by palps and carapace lobes) ground level under stones and in vegetable detritus; several species have been taken at the snow line. A few have been found in ants’ nests. Descriptions of the species.-The species are described in the order shown in the list given earlier. All figures of palps are of the right palp. The holotypes of the new species are deposited in AMNH, MCZ or CNC, as given under the species description. Scotinotylus eutypus (Chamberlin), new combination Figures 16, 18, 19, 20, 21, 24; Map 3 Ceratinopsis eutypa Chamberlin 1948:509 (female) Scotinotylus antennatus: Bishop and Crosby 1938:55 (male) (not S. antennatus (Cambridge)); Bonnet 1958:3962 (in part); Roewer 1942:686 (in part) MILLIDGE-GENUS SCOTINOTYLUS 111 Table 2.-Paitial key to Scotinotylus species: females. 1. Epigynum with cusp-like knob arising from the anterior margin (e.g. Fig. 20); palpal tibia with 2 trichobothria S. eutypus, S. alpinus, S. alienus, S. protervus (separate by epigyna) 2. Epigynum with tongue-like process arising from the anterior region; palpal tibia with 2 tricho- bothria a. process short, standing more or less erect from the anterior margin (Figs. 27, 31) i. clear epigastric striae present S. pallidus ii. epigastric striae absent or very faint S. sacer, S. sacratus (see species descriptions) b. process longer, more or less prone i. process long and broad (Figs. 85, 91) S. majesticus, S. magnificus, S. regalis (see species descriptions) ii. process short and broad (Figs. 77, 80) S. vernalis, S. exsectoides (separate by epigyna) iii. process long and narrow (Figs. 61, 65) S. patellatus, S. gracilis (separate by epigyna) 3. Epigynum not as 1. or 2., but with 2 “cusps” anteriorly (e.g. Figs. 122-132); palpal tibia usually with 3 trichobothria, one of which may be very small a. tibial spines 1111; palpal tibia with 2 trichobothria S. formicarius b. tibial spines 2221 i. epigynum with points of cusps directed forwards (Figs. 131, 132) S. bodenburgi, S. sagittatus (separate by epigyna) ii. epigynum with points of cusps directed backwards (e.g. Fig. 122) Epigynum as Fig. 102 S. pollucis Epigynum of general form shown in Figs. 95, 122 S. kenus, S. castoris, S. sanctus, S. crinitus, S. montanus, S. bicavatus, S. bipoculatus, S. boreus, S. sintalutus, S. dubiosus, S. apache (see species descriptions) 4. Epigynum with neither cusps nor tongue-like process a. epigynum as Fig. 151 S. monoceros, S. bicornis (see species descriptions) b. epigynum as Fig. 121 S. ambiguus Type. —Female type from Rainier Park, Washington, August 9, 1929 (R. V. Cham- berlin). Description.— Chamberlin’s specimens (including the type) of Ceratinopsis eutypa can- not be traced, but the epigynum (Chamberlin 1948: Fig. 36) agrees well with that of the specimen here described; the size quoted is also practically the same. The type locality of C. eutypa corresponds with the locality of the specimens described here. The male and female were taken together. Total length: female 2.0 mm, male 1.9 mm. Carapace: length: female 0.80 mm, male 0.85 mm. Orange-brown, with blackish markings and margins. The male carapace is raised into a small lobe anteriorly, and the clypeus projects distinctly (Fig. 16); there are 2 short curved horn-like spines in the ocular area. Abdomen: grey-black, epigastric plates smooth. Sternum: black. Legs: orange-brown. Tibial spines: female 2221, male 0021. Tml: female 0.36-0.38, male 0.40-0.45. Male palp: Figs. 18, 19, 21 ; the embolus is long and thin distally, and the tibia bears one stout 178 THE JOURNAL OF ARACHNOLOGY spine dorsally. Female palp: tibia with 2 trichobothria. Epigynum: Fig. 20; heavily pig- mented. The internal genitalia are probably very close to those of S. antennatus (Fig. 23). Diagnosis.-This species closely resembles the European species S. antennatus , with which it was confused by Bishop and Crosby (1938). The two forms are here regarded as separate species because of the existence of small structural differences, reinforced by the probability that the North American and European populations have been genetically isolated from one another for a considerable period of geological time. The palpal organs and the tibial apophyses of the two species are virtually identical, but S. eutypus shows differences in the shape of the male carapace, and the horn-like spines are shorter (Fig. 16 cf. Fig. 17). The females of the two species are closely similar, and since only one female of S. eutypus has been examined it is uncertain whether the epigyna are distinguishable (Fig. 20 cf. Fig. 22). So far as the North American fauna is concerned, the male of S. eutypus can be diagnosed immediately by the form of the carapace, and confirmation is given by the form of the palp and palpal tibia (Figs. 18, 19). The female is diagnosed by the epigynum, which has a small dark colored cusp-shaped knob on the anterior margin, a character which groups this species with S. alienus, S. alpinus and S. protervus\ from these three species, S. eutypus is distinguished without difficulty by the form of the epigynum (Fig. 20 cf. Figs. 47, 55, 72). 27 29 Figs. 25-31.-5. sacer : 25, male palp, ectal; 26, male palp, meso-ventral; 27, epigynum; 28, male carapace, lateral; 29, internal genitalia, female, ventral; 30, male palpal tibia, ectal; 31, epigynum, lateral. Abbreviation: M, membraneous part of suprategular apophysis (Scale lines 0.1 mm). MILLIDGE-GENUS SCOTINOTYLUS 179 Distribution.— This species is known from Washington, Oregon and British Columbia (Map 3); I have seen only specimens from Washington. Natural History.— The species has been taken from near the snow line in Washington. Males and females have occurred in August and September. Scotinotylus sacer (Crosby), new combination Figures 25, 26, 27, 28, 29, 30, 31 ; Map 1 Cochlemholus sacer Crosby 1929:82; Roewer 1942:660; Bonnet 1956:1175; Holm 1967:14 Lophocarenum alpinum: Emerton 1915: 150 (male only; not Dismodicus alpinus Banks) Type.— Male holotype from Lake Louise, Alberta, August 4, 1927; in AMNH. The species has been so well described in the past (references above) that examination of the type was not considered to be necessary. Description.— Total length: female 1.65-1.8 mm, male 1.7-1. 8 mm. Carapace: length: female/male 0.75-0.80 mm. Deep brown to orange-brown, with faint dusky markings. The male has a rather shallow lobe, with a weak longitudinal furrow, and the clypeus projects (Fig. 28). Abdomen: pale grey to grey; epigastric plates smooth in both sexes. Sternum: orange, reticulated with black. Legs: orange-brown. Tibial spines: female/male 2221, but weak on legs I and II in male. Tml: female 0.36-0.40, male 0.40. Male palp: Figs. 25, 26, 30; the tibia has 3 stout spines dorsally, and there is a distinct knob on the ecto-dorsal side. Female palp: tibia with 2 trichobothria. Epigynum: Figs. 27, 31; the degree of pigmentation is variable; the markings within the posterior plate, though some- what variable, always have the wavy outline shown. Internal genitalia: Fig. 29. Diagnosis.— The male of S. sacer is grouped with S. alpinus, S. pallidus and S. sacratus by the single carapace lobe and the presence on the palpal tibia of 2-3 stout spines. From S. alpinus, S. sacer is separated by the form of the carapace, the lobe being considerably larger in S. alpinus (Fig. 28 cf. Fig. 53), by the palpal tibia (Figs. 25, 30 cf. Figs. 51 , 54) and by the shorter and stouter embolus of S. alpinus (Fig. 26 cf. Fig. 52). S. sacer male is distinguished from S. pallidus by its normally larger size, by the somewhat greater projec- tion of the clypeal region (Fig. 28 cf. Fig. 35), by the form of the palpal tibia which in S. pallidus lacks the ecto-dorsal knob present in S. sacer (Fig. 30 cf. Fig. 34), and by the presence in S. pallidus of clear epigastric striae. S. sacratus male has a larger carapace lobe than in S. sacer (Fig. 28 cf. Fig. 39), the lobe being somewhat similar to that of S. alpinus, and the palpal tibia of S. sacratus lacks the ecto-dorsal knob present in S. sacer (Fig. 30 cf. Fig. 38). The female of S. sacer is diagnosed by the epigynum, which has a short tongue-like process standing more or less erect from the anterior margin; this character groups S. sacer with S. sacratus and S. pallidus. The females of S. sacer and S. sacratus may be distinguishable by the greater length of the tongue in S. sacer (Fig. 31 cf. Fig. 43), but more specimens of S. sacratus are required for confirmation of this dif- ference. The females of S. sacer and of S. sacratus are separable from S. pallidus by their somewhat larger size, and by the presence in S. pallidus of clear epigastric striae; the epigyna of these three species are very similar in form (Figs. 27, 36, 41), but that of S. pallidus is smaller in size and less strongly marked. Distribution.— S. sacer has been recorded from Alaska, Yukon Territory, British Columbia, Mackenzie, Alberta, Wyoming and Oregon; it is also known from west Green- land (Holm 1967) (Map 1). 180 THE JOURNAL OF ARACHNOLOGY Natural History.— Males have been taken in April, June and August, females in April, June, August and September; the chief maturity period is probably in late spring and summer. Holm (1967) records that the species was mostly caught (in west Greenland) in pitfall traps and by sifting leaf litter in herbaceous areas, heathland and bog. Scotinotylus pallidus (Emerton), new combination Figures 32, 33, 34, 35, 36, 37; Map 3 Lophocarenum pallidum Emerton 1882:480; 1909:176 Cochlembolus pallidus : Crosby and Bishop 1933:168; Roewer 1942:660; Bonnet 1956:1175 Erigone pallens Marx 1890:5 35 (not Erigone pallens Cambridge 1872) Type.— The type locality was reported to be White Mountains, New Hampshire, and the date of capture June 1878. The type material in MCZ is labelled “from glen and base of Mt. Washington, June 1877”; this vial contains one female of S. pallidus , one male of an unidentified species and females of two unidentified species. A second vial of ? type material from Mt. Washington, N.H., June 1877, contains one male of S. pallidus , and a male and female of another species. At that period, type material was frequently regarded as of little importance, and this mixture of species in the type vials is not particularly surprising. Description.-Total length: female 1.1- 1.7 mm. male 1.1-1.65 mm. Carapace: length: female 0.62-0.70 mm, male 0.65-0.70 mm. Pale brown to orange-brown, with dusky or black markings and margins. The male carapace is raised into a rather shallow lobe and the clypeus projects (Figs. 33, 35). Abdomen: pale grey to black; the epigastric plates are striated, with the striae fairly closely spaced in the female, and fairly widely spaced in the male. Sternum: yellow with dusky margins in the paler specimens, but almost black in the Figs. 32-31. -S. pallidus : 32, male palp, mesal; 33, male carapace, dorsal; 34, male palpal tibia, ectal; 35, male carapace, lateral; 36, epigynum; 37, internal genitalia, female, ventral (Scale lines 0.1 mm). MILLIDGE-GENUS SCOTINOTYLUS 181 darker specimens. Legs: pale orange-brown to brown. Tibial spines: female 2221, male 0021 or 1121, but very weak. Tml: female/male 0.37-0.40. Male palp: Figs. 32, 34; the tibia bears 2-3 stout spines. Female palp: tibia with 2 trichobothria. Epigynum: Fig. 36; internal genitalia Fig. 37. The specimens of this species from the north-eastern United States, Ontario and Quebec are small and pale in color, but specimens presumably of this species (having identical palps and epigyna) from further west and south are larger and usually much darker in color. Diagnosis.— S. pallidus is closely related to S. sacer, and its diagnosis is dealt with under that species. Distribution.— S. pallidus has a wide distribution, having been recorded from Ontario, Manitoba, Quebec, New Brunswick, Saskatchewan, Alberta, New York, New Hampshire, Massachusetts, Connecticut, Michigan, Montana, Utah, S. Dakota, Colorado, Wyoming and New Mexico (Map 3). Natural History.— Males have been taken in March-October, females in practically every month of the year. In Canada it has been taken in pitfall traps in woods and in a bog; there is no information on habitats in U.S.A. Scotinotylus sacratus , new species Figures 38, 39, 40, 41, 42, 43; Map 3 Type.— Male holotype from Mirror Lake, Uintah Mts., Utah, July 28, 1936 (W. Ivie); deposited in AMNH. Description.— The two sexes were taken together. Total length: female 1.75-1.85 mm, male 1.65-1.75 mm. Carapace: length: female/male 0.80 mm. Orange-brown with dusky markings and margins. Male carapace with large lobe, which is rather triangular in outline when viewed dorsally (Figs. 39, 40). Abdomen: grey to black; epigastric plates smooth or with very weak striae. Sternum: orange-brown, suffused with black. Legs: orange-brown. Tibial spines: female/male 2221, but thin and weak in the male. Tml: female/male Figs. 38-43. -S'. sacratus : 38, male palpal tibia, ectal; 39, male carapace, lateral; 40, male carapace, dorsal; 41, epigynum; 42, male palpal tibia, dorsal; 43, epigynum, lateral (Scale lines 0.1 mm). 182 THE JOURNAL OF ARACHNOLOGY 0.37-0.40. Male palp: the palpal organs are practically identical with those of S. sacer. The tibia has 3 stout spines; the tibial apophysis (Figs. 38, 42) lacks the ecto-dorsal knob present in S. sacer. Female palp: tibia with 2 trichobothria. Epigynum: Figs. 41, 43. Diagnosis.— S. sacratus is closely related to S. sacer, and its diagnosis is dealt with under that species. Distribution.— This species is known only from Utah and Colorado (Map 3). Natural History.— Males have been taken in July, females in July and August. One female was caught in spruce, fir forest at 3535 m in Colorado. Scotinotylus alienus (Kulczynski), new combination Figure 47; Map 1 Erigone (Ceratinopsisl) aliena Kulczynski 1885:40 (male) Caledonia aliena: Hull 1911:49; Roewer 1942:659 Caledonia evansi : Holm 1960:111 (probably) Ceratinopsis aliena : Bonnet 1956:1017 Type.-Male type from Kamchatka, eastern Siberia. This is probably in Instytut Zoologiczny at Warsaw, Poland, but I have not been able to borrow it for examination. I have followed Holm (1960) in accepting that S. alienus is probably not identical with S. evansi (Cambridge). If this assumption is correct, the species found in western North America is most likely to be S. alienus, which was described from material taken in an adjacent area in eastern Asia. An analogous case is the presence of Walckenaeria ( Corni - cularia ) lepida (Kulczynski) in both Kamchatka and western North America (recorded as its synonym Cornicularia pacifica Emerton: Ivie 1965). I have assumed therefore that the females taken in north-western America are S. alienus, but capture of the male is neces- sary to confirm (or refute) this view. Description.— Total length: female 2.35 mm. Carapace: length: female 0.90-1.0 mm. Chestnut-brown with dusky markings and margins. Abdomen: grey; epigastric plates smooth. Sternum: brown, heavily suffused with black. Legs: brown to orange-brown. Tibial spines: female 2221. Tml: female 0.48-0.50. Female palp: tibia with 2 tricho- bothria. Epigynum: Fig. 47; the internal genitalia are probably almost identical with those of S. evansi (Fig. 48). According to Kulczynski’s description and figures (1885), the male of S. alienus is generally similar to S. evansi, but there is a difference in the position of the pointed apophysis which projects from the first turn of the embolic coil. Diagnosis. -Kulczynski’s description shows that the male of S. alienus is very similar to S. evansi, with the carapace of the same form (Fig. 46). The male will therefore fall into Section 2 of the Key, with S. protervus, from which it will readily be separated by the form of the carapace (Fig. 46 cf. Fig. 71). The female of S. alienus is diagnosed by the form of the epigynum, which has a dark colored cusp or small knob on the anterior margin, a character which groups it with S. eutypus, S. alpinus and S. protervus. From these three species, S. alienus is distinguished without difficulty by the form of the epigynum (Fig. 47 cf. Figs. 20, 55, 72). The epigynum of S . alienus seems to be indistin- guishable from that of the European species S. evansi. Distribution.— This species is known in North America from Alaska ( Caledonia evansi : Holm 1960), Alberta and Yukon Territory (Map 1). Natural History.— The females were taken in June and August. Nothing is recorded on habitat. MILLIDGE- GENUS SCOTJNOTYLUS 183 Scotinotylus alpinus (Banks), new combination Figures 3, 51, 52, 53, 54, 55, 56, 57; Map 2 Dismodicus alpinus Banks 1896:63 Gongylidium lapidicola Soerensen 1898:204 (female, not male) Gonatium inflatum. Soerensen 1898:206 (male) Lophocarenum alpinum: Emerton 1909:190 Tortembolus alpinus : Crosby 1925:115 Cochlembolus alpinus: Crosby 1929:79; Roewer 1942:660; Bonnet 1956:1175; Holm 1967:12 Scotynotylus ungavensis Jackson 1933:150 Coryphaeolana lapidicola : Braendegaard 1937:10 (female, not male); 1946:45. Scotynotylus lapidicola: Holm 1958:526 not Lophocarenum alpinum (male): Emerton 1915:150 Type.— Type from Mount Washington, New Hampshire; in MCZ, examined. Figs. 44-50.— 44, S. evansi , male palp, ectal; 45, S. evansi, male palp, mesal; 46, S. evansi, male carapace, lateral; 47, S. alienus, epigynum; 48, S. evansi, internal genitalia; female, ventral; 49, S. evansi, male palpal tibia, dorsal; 50, S. alpigena, epigynum. Abbreviations: E, embolus; M, membraneous part of suprategular apophysis; O, opening of spermathecal duct. (Scale lines 0.1 mm). 184 THE JOURNAL OF ARACHNOLOGY Description.— Total length: female 2.0-2.65 mm, male 1.8-2.05 mm. Carapace: length: female 0.90-1.10 mm, male 0.90-1.0 mm. Orange-brown to chestnut-brown, with dusky markings. Male carapace with a large lobe and projecting clypeus (Fig. 53). Abdomen: grey to black; epigastric plates smooth. Sternum: orange-brown, suffused with black. Legs: brown to orange-brown. Tibial spines: female/male 2221, but very weak in male on legs I and II. Tml: female 0.45-0.50, male 0.43-0.50. Male palp: Figs. 51, 52, 54; the embolus is fairly short and stout distally, and the tibia has 3 stout spines. Female palp: tibia with 2 trichobothria. Epigynum: Figs. 55, 57; the degree of pigmentation is variable, but the very dark form seems to be the commonest. Internal genitalia: Fig. 56. Diagnosis.— The male of S. alpinus is grouped with S. sacer, S. sacratus and S. pallidus by the single carapace lobe and the presence on the palpal tibia of 2-3 stout spines; its separation from these three species is dealt with under S. sacer diagnosis. The female of S. alpinus is diagnosed by the epigynum, which has a cusp-shaped knob on the anterior margin, a character which groups the species with S. eutypus, S. alienus and S. protervus. From these three species, S. alpinus is distinguished without difficulty by the form of the epigynum (Figs. 55, 57 cf. Figs. 20, 47, 72); it must be borne in mind that the depth of pigmentation of the epigynum varies considerably. Figs. 51-57. -S. alpinus : 51, male palp, ectal; 52, male palp, mesal; 53, male carapace, lateral; 54, male palpal tibia, dorsal; 55, epigynum; 56, internal genitalia, female, ventral; 57, epigynum, pale specimen. Abbreviation: M, membraneous part of suprategular apophysis. (Scale lines 0.1 mm). MILLIDGE-GENUS SCOTINOTYLUS 185 Distribution.— This species has a wide distribution in the northern parts of the con- tinent. It has been recorded from Alaska, Yukon Territory, British Columbia, Alberta, Manitoba, Quebec, Baffin Island and New Hampshire, but also from high altitudes in Wyoming and Colorado (Map 2). It is also known from Greenland (Holm 1967). Natural History. -Males and females have been taken from June to September; the maturity period is in summer. Holm (1967) states that the species is found almost exclusively under stones. Scotinotylus patellatus (Emerton), new combination Figures 58, 59, 60, 61 , 62, 63, 64; Map 3 Araeoncus patellatus Emerton 1917:262; Roewer 1942:685 (the locality quoted should be “British Columbia,” not “Columbia”); Bonnet 1955:378 Type.— Male holotype from Metlakatla, British Columbia (J. H. Keen): in MCZ, examined. Since 1903, this locality has been in Alaska. Description.— Total length: female 1.65-1.90, male 1.7 mm. Carapace: length: female 0. 8-0.9 mm, male 0.85-0.90 mm. Orange-brown, with dusky markings. Male carapace not raised into lobe (Fig. 60); numerous long backward-directed bristles arise from the ocular region. Abdomen: grey to black; epigastric plates smooth. Sternum: orange, suffused with black. Legs: orange-brown. Tibial spines: female 2221, male 0021. The curved hairs on tibiae I and metatarsi I of the male are strongly developed. Tml: female 0.46-0.53, male 0.48. Male palp: Figs. 58, 59, 64; the tibia bears 2-3 stout spines, and the femur is somewhat swollen. Female palp: tibia with 2 trichobothria. Epigynum: Figs. 61, 63; the tongue is always long, but the width shows small variations. Internal genitalia: Fig. 62. Diagnosis.— The male of S. patellatus is diagnosed by the absence of a raised lobe on the carapace, a character which groups it with S. protervus and S. alienus. From these two species it is distinguished by the presence of 3 fairly stout spines on the palpal tibia (Fig. 58), by the form of the carapace (Fig. 60 cf. Figs. 46, 71) and by the form of the embolus. The embolus is simple in S. patellatus (Figs. 58, 59) but has a terminal loop in S. protervus (Figs. 67, 68), while in S. alienus there is a pointed apophysis arising from the first turn of the embolic coil similar to that in S. evansi (Fig. 45). The female epigynum of S. patellatus has a very distinctive club-shaped “tongue” arising from the anterior margin (Figs. 61, 63), which permits immediate diagnosis of the species. S. gracilis has a somewhat similar epigynal process, which is however longer and much more slender (Fig. 65). Distribution.— This species appears to be widespread along the western coastal area; it has been recorded from Alaska, British Columbia, Washington, Oregon and California (Map 3). Natural History.— The male has been taken in January-February, the female in March, May, July, August, October and December. Nothing is recorded on habitat. Scotinotylus gracilis, new species Figures 65, 66; Map 3 Type. —Female holotype from west of Inverness, Marin Co., California, November 8, 1953 (V. Roth and G. Marsh); deposited in AMNH. 186 THE JOURNAL OF ARACHNOLOGY Description.— Only the female is known. Total length: female 1.9 mm. Carapace: length: female 0.80 mm. Orange-brown, with black markings and margins. Abdomen: black; epigastric plates smooth. Sternum: black. Legs: orange-brown. Tibial spines: female 2221. Tml: female 0.47. Female palp: tibia with 2 trichobothria. Epigynum: Fig. 65; there is a long, thin, semi-transparent tongue arising from the anterior margin. Internal genitalia: Fig. 66. Diagnosis.— The species is diagnosed by the form of the epigynum, which has a long slender process arising from the anterior margin (Fig. 65); this process is longer, narrower and more transparent than the corresponding process in 5. patellatus (Fig. 61). Distribution.— Known only from the type locality (Map 3). Natural History.— The female was taken in November, but nothing was recorded on habitat. M Figs. 58-66.-58, S. patellatus, male palp, ectal; 59, S. patellatus, male palp, mesal; 60, S. patel- latus, male carapace, lateral; 61,5. patellatus, epigynum, ventral; 62, S. patellatus, internal genitalia, female, ventral; 63, S. patellatus, epigynum, lateral; 64, S. patellatus, male palpal tibia, dorsal; 65, S. gracilis, epigynum; 66, S. gracilis, internal genitalia, female, ventral. Abbreviation: M, membraneous part of suprategular apophysis. (Scale lines 0.1 mm). MILLIDGE-GENUS SCOTINOTYLUS 187 Scotinotylus protervus (L. Koch), new combination Figures 5, 67, 68, 69, 70, 71, 72, 73; Map 1 Erigone proterva L. Koch 1879:70; Bonnet 1956:1772 Caledonia proterva : Holm 1960:112; 1970:190 Type.— Female holotype from Tunguska, west Siberia; in Naturhistoriska Riksmuseum, Stockholm. The female has such a characteristic epigynum (Holm 1960) that examination of the type was considered to be unnecessary. Description.— The only previous description of the male (Holm 1970) was based on a specimen which had not completed its final moult. Total length: female 2. 5-2.9 mm, male 2.7 mm. Carapace: length: female 1.1-1.25 mm, male 1.2 mm. Brown to chestnut-brown. The male carapace is raised anteriorly, the elevation bearing numerous bristles (Fig. 71); the clypeus projects strongly. Abdomen: grey to black; the epigastric plates are smooth. Sternum: orange-brown, suffused with black. Legs: brown to orange-brown. Tibial spines: female/male 2221, but short and weak in the male. Tml: female 0.65-0.70. male 0.67. Figs. 67-73.-5'. protervus: 67, male palp, ectal; 68, male palp, mesal; 69, male palpal tibia, dorsal; 70, male palpal tibia, ecto-dorsal; 71, male carapace, lateral; 72, epigynum; 73, internal genitalia, female, ventral. Abbreviation: T, tegulum. (Scale lines 0.1 mm). 188 THE JOURNAL OF ARACHNOLOGY Male palp: Figs. 67, 68, 69, 70; the embolus ends in a small loop. There are no stout spines on the palpal tibia. Female palp: tibia with 2 trichobothria. Epigynum: Fig. 72; the internal ducts follow a rather sinuous course (Fig. 73). Diagnosis.— S. protervus male is diagnosed by the distinctive form of the carapace (Fig. 71), by the absence of stout spines on the palpal tibia, and by the form of the embolus, which terminates in a small loop (Figs. 67, 68). The female is diagnosed by the epigynum, which has a small cusp-shaped knob on the anterior margin, in common with S. eutypus, S. alpinus and S. alienus ; from these three species it is readily separated by the distinctive form of the epigynum, which has large, clear cavities on either side of the median septum (Fig. 72, cf. Figs. 20, 47, 55). Distribution.— In North America this species is known only from Alaska and Yukon Territory (Map 1); Holm’s records (Holm 1960, 1970) are included on this map. Natural History. -Adult males have been taken in August; Holm (1970) recorded a male near its final moult in July. The female has been taken in June, July and August. In Alaska, both sexes were taken together under rocks on tundra at 1200 m. Scotinotylus vernalis (Emerton), new combination Figures 6, 74, 75, 76, 77, 78, 79; Map 2 Lophocarenum vernalis Emerton 1882:51 Erigone vernalis : Marx 1889:536 Diplocephalus vernalis : Banks 1910:27 Cochlembolus vernalis : Crosby 1929:82; Roewer 1942:660; Bonnet 1956:1175 Scylaceus divisus Chamberlin 1948:544 NEW SYNONYMY. The type female (AMNH) has been examined and found to be identical with S. vernalis. Type.— Male and female types from Pine Rock, New Haven, Connecticut, in March; in MCZ, examined. Description.— Total length: female 1.8-1.85 mm, male 1 .75-1 .9 mm. Carapace: length: female 0.75 mm, male 0.85-0.9 mm. Orange-brown, with faint dusky markings. Male carapace raised into a small lobe (Fig. 76). Abdomen: grey. Epigastric plates with weak striae in male; striae absent or extremely weak in female. Sternum: orange, suffused with black. Legs: orange. Tibial spines: female 2221, male 0211. Tml: female 0.41-0.46, male 0.42. Male palp: Figs. 74, 75, 78; there are no stout spines on the tibia. Female palp: tibia with 2 trichobothria. Epigynum: Fig. 77; brown, suffused with black. There is a broad tongue-shaped process arising from the anterior area. Internal genitalia: Fig. 79. Diagnosis.— The male of S. vernalis is diagnosed by the presence of a single lobe on the carapace, and by the absence of stout spines on the palpal tibia; these characters place it with S. regalis and S. pollucis. From these two species S. vernalis is distinguished by its shallower lobe (Fig. 76 cf. Figs. 89, 99) and by the form of the palpal tibia, which has two small pointed apophyses (Figs. 75, 78 cf. Figs. 84, 101). The female of S. vernalis is diagnosed by the epigynum, which has a broad tongue-like process (Fig. 77); the only species with which it could be confused is S. exsectoides, which has a relatively smaller tongue and is paler in color (Fig. 77, cf. Fig. 80). Distribution.— This species is known from Ontario, Michigan, Massachusetts, Con- necticut, Nebraska and N. Dakota (Map 2) Natural History.— Most of the records are for males, females having been taken on very few occasions; the reason for this is not known. Males have been taken in January, April, May-June, October, November and December, females only in April. Nothing is recorded on habitat. MILLIDGE-GENUS SCOTINOTYLUS 189 Scotinotylus exsectoides, new species Figures 80, 81 ; Map 2 Type .—Female holotype from Actinolite, Ontario, April 25, 1966 (G. Ayre); deposited in CNC, Ottawa. Description. -Only the female is known. Total length: female 2.0-2.4 mm. Carapace: length: female 0.90-0.95 mm. Orange-brown with faint dusky markings. Abdomen: grey; striae on epigastric plates very weak or absent. Sternum: orange, with dusky margins. Legs: orange to orange-brown. Tibial spines: female 2221. Tml: female 0.51-0.54. Female palp: tibia with 2 trichobothria. Epigynum: Fig. 80; pale orange or yellow in color. There is a broad tongue-shaped process arising from the anterior area. Internal genitalia: Fig. 81. Diagnosis. -The female of this species is diagnosed by the epigynum (Fig. 80), which like that of S. vernalis (Fig. 77) has a broad tongue-like process. The tongue in S. exsectoides is relatively shorter, the epigynum is paler in color, and there are differences in the structural detail. Distribution.— Known only from the type locality (Map 2). Natural History. -Numerous females were taken in April at the type locality, in the brood nest of the ant Formica exsectoides Forel. Figs. 74-81.-74, S. vernalis , male palp, meso-ventral; 75, S. vernalis , male palpal tibia, dorsal; 76, S. vernalis, male carapace, lateral; 77, S. vernalis, epigynum; 78, S. vernalis, male palpal tibia, ectal; 79, S. vernalis, internal genitalia, female, ventral; 80, S. exsectoides, epigynum; 81, S', exsectoides, internal genitalia, female, ventral. (Scale lines 0.1 mm). 190 THE JOURNAL OF ARACHNOLOGY Scotinotylus majesticus (Chamberlin and Ivie), new combination Figures 4, 14, 82, 83, 84, 85, 86, 87; Map 3 Yukon majesticum Chamberlin and Ivie 1947:52 Type.-Male holotype from Matanuska Valley, Alaska, August 23-31, 1943 (J. C. Chamberlin); in AMNH. Paratypes (AMNH) examined. Description.— Total length: female: 2.55-2.75 mm, male 2.30-2.35 mm. Carapace: length: female 1.10-1.15 mm, male 1.05-1.10 mm. Orange-brown, with dusky markings and margins. Male carapace with 2 lobes, one large, one small (Fig. 86); the size and shape of the lobes shows small variations. Abdomen: grey to black; striae on epigastric plates very weak or absent in female, weak and closely spaced in male. Sternum: orange-brown, heavily suffused with black. Legs: orange-brown. Tibial spines: female 2221, male 0221 but short and weak. Tml: female/male 0.57-0.60. Male palp: Figs. 82, 83, 84; there are no stout spines on the tibia. Female palp: tibia with 2 trichobothria. Epigynum: Fig. 85; the length and width of the stout tongue-shaped process shows some variation. The internal ducts follow a very sinuous course (Fig. 87). 87 Figs. 82-87. -S'. majesticus : 82, male palp, ectal; 83, male palp, mesal; 84, male palpal tibia, dorsal; 85, epigynum; 86, male carapace, lateral; 87, internal genitalia, female, ventral. Abbreviation: M, membraneous part of suprategular apophysis. (Scale lines 0.1 mm). MILLIDGE-GENUS SCOTINOTYLUS 191 Diagnosis.— The male of S. majesticus is diagnosed by the form of the carapace (Fig. 86), which at once separates it from all the other species in the genus; confirmation is afforded by the palp (Figs. 82, 83) and palpal tibia (Fig. 84). The female is diagnosed by the epigynum, which has a long broad process arising from the anterior margin (Fig. 85); this character separates S. majesticus from all other species in the genus except S. magnificus and S, regalis. The females of S. majesticus and S. magnificus appear to be indistinguishable by the epigynum or any other character. The epigynum of S. regalis is very similar to that of S. majesticus , but the two species are distinguishable by the form of the posterior wing-like markings (Fig. 85 cf. Fig. 91). Distribution.— This species is known from Alaska, Yukon Territory, Colorado and Wyoming (Map 3). Females taken without males in Colorado may be either S. majesticus or S. magnificus and are not recorded on Map 3. Natural History.— Males have been taken in July and August, females in July, August, September and October. In Colorado and Wyoming it is found only at high altitudes (3-4000 m). Nothing is recorded on habitat. Scotinotylus magnificus , new species Figure 88; Map 3 Type. —Male holotype from Independence Pass, 3700 m, Sawatch Mts., Lake Co., Colorado, July 21, 1961 (H. and L. Levi); deposited in MCZ. Description.— The male and female were taken together. Total length: female 2.3-2.55 mm, male 2.0 mm. Carapace: length: female 1. 0-1.1 mm male 0.95 mm. Orange-brown, with blackish markings and margins. Male carapace raised into a large lobe, with a tiny additional lobe anterior to it (Fig. 88); no intermediates between this carapace form and that of S. majesticus have been seen. Abdomen: grey to black; striae on epigastric plates Figs. 88-91.-88, S. magnificus, male carapace, lateral; 89, S. regalis, male carapace, lateral; 90, S. regalis, male palp, mesal; 91, S. regalis, epigynum. Abbreviation: L, lamellar apophysis from embolus. (Scale lines 0.1 mm). 192 THE JOURNAL OF ARACHNOLOGY very weak or absent in female, weak and closely spaced in male. Sternum: orange-brown, heavily suffused with black. Legs: brown to orange-brown. Tibial spines: female/male 2221, but very short in male and sometimes absent on tibiae I and II. Tml: female 0.52-0.55, male 0.54-0.57. Male palp: identical with that of S. majesticus. Female palp: tibia with 2 trichobothria. Epigynum: not distinguishable from that of S. majesticus. Diagnosis. -S. magnificus male is diagnosed by the form of the carapace, which has one large and one tiny lobe (Fig. 88); confirmation is given by the palp, which is devoid of stout spines and is not distinguishable from that of S. majesticus. The females of S. magnificus and S. majesticus appear to be structurally indistinguishable. Distribution.— This species is known only from Colorado (Map 3). Natural History.— Both sexes were taken in July, at a high altitude (3700 m) in Colorado; nothing was recorded on habitat. Scotinotylus regalis, new species Figures 89, 90, 91 ; Map 3 Type.— Male holotype from Last Chance Gulch, Helena, Jefferson Co., Montana, October 3, 1964 (J. and W. Ivie); deposited in AMNH. Figs. 92-91. -S. kenus : 92, male palp, ectal; 93, male palpal tibia, dorsal; 94, male palp, mesal; 95, epigynum; 96, male carapace, lateral; 97, epigynum, a paler specimen. Abbreviation: M, membraneous part of suprategular apophysis. (Scale lines 0.1 mm). MILLIDGE-GENUS SCOTINOTYLUS 193 Description. -The male and female were taken together. Total length: female 2.30-2.60 mm, male 2.0-2. 1 mm. Carapace: length: female 1.0-1. 2 mm, male 1.0 mm. Orange-brown, with dusky markings and margins. Male carapace raised into a single lobe, which bears long, forward-directed bristles anteriorly (Fig. 89). Abdomen: grey to black; striae on epigastric plates very weak in female, weak and closely spaced in male. Sternum: orange, heavily suffused with black. Legs: orange to orange-brown. Tibial spines: female/ male 2221, but very short in male and sometimes absent on tibiae I and II. Tml: female/ male 0.50-0.53. Male palp: Fig. 90; arising from near the base of the embolic coil there is a lamellar apophysis (L, Fig. 90), but apart from this, the palp is closely similar to that of S. majesticus/S. magnificus. Female palp: tibia with 2 trichobothria. Epigynum: Fig. 91; the length of the tongue shows small variation. Figs. 98-107.-98, S. pollucis, male palp, ectal; 99, S. pollucis, male carapace, lateral; 100, S. pollucis, male palp, mesal; 101, S. pollucis, male palpal tibia, dorsal; 102,5. pollucis, epigynum; 103, 5. pollucis, tibia and metatarsus of leg I, male; 104, S. crinitus, male carapace, dorsal; 105, 5. crinitus, male palpal tibia, dorsal; 106, 5. crinitus, male carapace, lateral; 107,5. crinitus, epigynum. (Scale lines 0.1 mm). 194 THE JOURNAL OF ARACHNOLOGY Diagnosis.— The male of S. regalis is diagnosed by the single large lobe on the carapace and by the absence of stout spines on the palpal tibia; these characters place it with S. pollucis in the Key. From this latter species, S. regalis is separated by its much wider embolic coil (Fig. 90 cf. Fig. 100), and by the form of the palpal tibia (Fig. 84 cf. Fig. 101). The palp of S. regalis is very similar to those of S. majesticus and S. magnificus , but is distinguished by the presence of the lamellar apophysis near the base of the embolus (L, Fig. 90). The female of S. regalis has the epigynum very similar to that of S. majesticus/ S. magnificus , but S. regalis is distinguishable by the form of the posterior wing-like markings (Fig. 91 cf. Fig. 85). Distribution.— This species is known only from the type locality (Map 3). Natural History.— Both sexes were taken in some number in October; nothing was recorded on habitat. Scotinotylus ambiguus, new species Figures 121 , 133; Map 3 Type.— Female holotype from Cedar Lake, N. Leadpoint, Washington, May 1962 (W. Ivie); deposited in AMNH. Description.— Only the female is known. Total length: female 1. 8-2.2 mm. Carapace: length: female 0.85-0.95 mm. Orange-brown. Abdomen: grey to black; epigastric plates with fairly clear striae. Sternum: orange, suffused with black. Legs: orange to brown. Tibial spines: female 2221. Tml: female 0.37-0.43. Female palp: tibia with 2 tricho- bothria. Epigynum: Fig. 121; internal genitalia Fig. 133. Diagnosis.— This species is diagnosed by the epigynum; the absence of a linguiform process and of cusps places it with S. monoceros in the Key. S. ambiguus is readily separable from S. monoceros by the form of the epigynum (Fig. 121 cf. Fig. 151). Distribution.— This species is recorded from British Columbia, Washington, Idaho and Wyoming (Map 3). Natural History.— Females were taken in March, May (numerous), June and July; nothing was recorded on habitat. Scotinotylus kenus (Chamberlin), new combination Figures 10, 11, 92, 93, 94, 95, 96, 97, 134; Map 4 Cheraira kena Chamberlin 1948:519. Note: it is assumed that the specific name kena was intended to be an adjective agreeing in gender with Cheraira. Type .—No type seems to have been designated by Chamberlin, but one vial in AMNH has a label which agrees with the details given by Chamberlin (1948) under “Type locality”: namely, Mirror Lake, Uintah Mts., Utah, July 28, 1936 (W. Ivie); 4 females. One of these females has been selected and labelled as “Lectotype,” and is deposited in AMNH. Description.— The male, described here for the first time, was not taken with a female, but as it came from the type locality (but on a different date) there can be little doubt as to its identity. Total length: female 2.2-2.7 mm, male 2.1 mm. Carapace: length: female/ male 1.0 mm; orange-brown, with dusky markings. The male carapace is raised into a large lobe (Fig. 96), which has a shallow longitudinal furrow; the clypeus projects MILLIDGE-GENUS SCOTINOTYLUS 195 strongly. Abdomen: grey; the epigastric plates have very weak, closely spaced striae in both sexes. Sternum: yellow to yellow-brown, with dusky margins. Legs: orange-brown to yellow-brown. Tibial spines: female 2221, male 1121 but weak. Tml: female 0.47, male 0.45-0.47. Male palp: Figs. 92, 93, 94; a slender black pointed apophysis arises from the first coil of the embolus. The palpal tibia bears one fairly stout spine. Female palp: tibia with 3 trichobothria, one rather small. Epigynum: Figs. 95, 97; internal genitalia Fig. 134. Diagnosis.— The male of S. kenus is diagnosed by the large lobe on the carapace, by the presence of the single stout spine on the palpal tibia and by the presence of a small black tooth on the anterio-ectal margin of the palpal tibia; these characters group S. kenus in the Key with S. sanctus, S. crinitus, S. humilis and S. montanus, and no doubt with other related species when the males of these are discovered. S. kenus is distinguished from S. humilis by the larger carapace lobe (Fig. 96 cf. Fig. 117) and by the much larger diameter of the embolic coil (Fig. 94 cf. Fig. 115). From S. sanctus, S. crinitus and S. montanus males it is readily distinguished by the form of the palpal tibia (Fig. 93 cf. Figs. 105, 112, 120), and by the presence on the embolic coil of the black needle-like apophysis (Fig. Figs. 108-1 14. -5. sanctus : 108, male palp, ectal; 109, male carapace, normal form, lateral; 110, male palp, meso-ventral; 111, male carapace, specimen from Arizona; 112, male palpal tibia, dorsal; 113, male carapace, anterior face, normal form; 114, male carapace, anterior face, specimen from Arizona. Abbreviations: M, membraneous part of suprategular apophysis; T, tegulum. (Scale lines 0.1 mm). 196 THE JOURNAL OF ARACHNOLOGY 94). The diagnosis of the female of S. kenus is based on the form of the epigynum, and is dealt with under S. sanctus diagnosis. Distribution.— This species is known from Montana, Utah and Arizona (Map 4). Natural History.— The male has been taken in August and October, the female in April, June and July. All the localities appear to be at moderately high altitudes, but nothing is recorded on habitat. Scotinotylus cast oris (Chamberlin), new combination Figure 122; Map 4 Cheraira castoris Chamberlin 1948:520 Type.— Holotype female from Beaver Canyon, 10 miles east of Beaver City, Utah, June 7, 1934 (W. Ivie and H. Rasmussen); in AMNH, examined. Description.— Only the female is known. Total length: female 2.4-2. 5 mm. Carapace: length: female 1. 0-1.1 mm; orange-brown with dusky margins. Abdomen: grey-black; epigastric plates smooth. Sternum: orange, faintly reticulated with black. Legs: orange. Tibial spines: female 2221. Tml: female 0.37-0.40. Female palp: tibia with 3 tricho- bothria. Epigynum: Fig. 122. Diagnosis.— S. castoris female is diagnosed by the epigynum; this is dealt with under S. sanctus diagnosis. Distribution.— Known only from the type locality (Map 4). Figs. 115-120.-115, S. humilis, male palpal organ, meso-ventral; 116, S. humilis, male palpal tibia, dorsal; 117, S. humilis , male carapace, lateral; 118, S. montanus, male palp, ectal; 119,5. montanus, male palp, meso-ventral; 120, S. montanus, male palpal tibia, dorsal. (Scale lines 0.1 mm). MILLIDGE-GENUS SCOTINOTYLUS 197 Natural History. -The two known females were taken in June; nothing was recorded on habitat. Scotinotylus pollucis , new species Figures 98, 99, 100, 101, 102, 103; Map 4 Type.-Holotype male from Rustlers Camp, Chiricahua Mts., Arizona, September 9, 1950 (W. J. Gertsch); deposited in AMNH. Description.— The male and female were taken together. Total length: female 1.9 mm, male 1.85-2.05 mm. Carapace: length: female 0.85-0.90 mm, male 0.90-0.95 mm. Orange-brown, with faint dusky markings. Male carapace raised into large lobe (Fig. 99) which has a weak longitudinal furrow. Abdomen: grey to black; striae on the epigastric plates are weak or absent. Sternum: yellow-brown, with dusky reticulations and margins. Figs. 121-132. -Epigyna. 121, 5. ambiguus ; 122, 5. castoris’, 123,5. sanctus', 124,5. montanus ; 125, 5. bicavatus ; 126, 5. bipoculatus ; 127, 5. boreus\ 128, 5. sintalutus ; 129, 5. dubiosus; 130, 5. apache ; 131,5. bodenburgi\ 132, 5. sagittatus. (Scale lines 0.1 mm). 198 THE JOURNAL OF ARACHNOLOGY Legs: orange-brown; leg I of the male has the tibia somewhat swollen, with ventral bristles, and the metatarsus is curved (Fig. 103). Tibial spines: female 2221, male 0121. Tml: female 0.40-0.44, male 0.41-0.45. Male palp: Figs. 98, 100, 101; the tibia lacks a stout spine. Female palp: tibia with 2 trichobothria, but sometimes a very small third one is also present. Epigynum: Fig. 102. Diagnosis.— The male of S. pollucis has a single large lobe on the carapace and no stout spines on the palpal tibia, which places it with S. regalis in the Key. S. pollucis male is readily separated from S. regalis by the much smaller diameter of the embolic coil (Fig. 100 cf. Fig. 90) and by the form of the tibial apophysis (Fig. 101 cf. Fig. 84); the membraneous parts of the SA are also quite different in form. The female of S. pollucis is diagnosed by the epigynum (Fig. 102), which is fairly readily distinguishable, by the shape of the posterior plate, from the numerous other species which have cusps on the epigynum (Section 3 of the Key). Distribution.— This species is known from several localities in Colorado and Arizona (Map 4). Natural History.— The male has been taken in September, the female in April, June, July, August and September. All the localities are at moderately high altitudes (2400-2800 m). Habitats recorded are in ponderosa and aspen at the side of a stream, in oak, juniper, douglas and aspen woods, and in meadow with aspen and lodgepole. Scotinotylus sanctus (Crosby), new combination Figures 8, 108, 109, 110, 111, 112, 113, 114, 123, 135; Map 4 Cochlembolus sanctus Crosby 1929:81 (male); Roewer 1942:660; Bonnet 1956:1175 Spirembolus chera Chamberlin and Ivie 1933:20 (female); Roewer 1942:665; Bonnet 1958:4122. NEW SYNONYMY: confirmed by examination of the type (AMNH). This species was later (Cham- berlin and Ivie 1945) synonymized, erroneously, with Disembolus stridulans Chamberlin and Ivie. Cheraira salmonis Chamberlin 1948:520 (female). NEW SYNONYMY: confirmed by examination of the type (AMNH). Type.— Holotype male from St. Johns, Utah, October 8, 1927 (R. V. Chamberlin); in AMNH, examined. Description.— The female has been taken with the male on a number of occasions in several different localities, and there can be no real doubt as to its identity. Total length: female 2.0-2. 1 mm, male 1. 9-2.0 mm. Carapace: length: female/male 0.90 mm. Brown to orange-brown, with dusky markings and margins. Male carapace raised into a moderately large lobe with a longitudinal furrow (Figs. 109, 113); the lobe is clothed anteriorly with fairly short hairs. A male from Arizona, with identical sex organs and accompanied by a typical female of S. sanctus , has the lobe significantly broader than normal (Figs. Ill, 114); it will be interesting to know if more specimens of this form turn up. Abdomen: grey; epigastric plates with weak striae or none in the female, but with clear, fairly widely spaced striae in the male. Sternum: brown, heavily suffused with black. Legs: orange- brown. Tibial spines: female 2221, male 0021, Tml: female 0.37-0.40, male 0.40-0.45. Male palp: Figs. 108, 110, 112; the tibia bears one stout spine. Female palp: tibia with 3 trichobothria, one small. Epigynum: Fig. 123; the distance apart and the size of the cusps show only small variations. Internal genitalia: Fig. 135. Diagnosis.— The male of S. sanctus has a single lobe on the carapace, a single stout spine on the palpal tibia, and a small black tooth on the anterio-ectal margin of the tibia. These characters place it with S. kenus, S. crinitus, S. humilis and S. montanus in the MILLIDGE-GENUS SCOTINOTYLUS 199 Key. S. sanctus male is easily separated from S. kenus by the form of the palpal tibia, which is much shorter in S. kenus (Fig. 1 12 cf. Fig. 93), and by the larger diameter of the embolic coil in S. kenus (Fig. 108 cf. Fig. 92). S. sanctus is distinguished from S. crinitus by the somewhat shorter palpal tibia (Fig. 1 12 cf. Fig. 105), and by the somewhat smaller diameter of the embolic coil, S. crinitus having the embolus very similar in size to that of S. montanus (Fig. 119). S. sanctus male is separated from S. humilis by the larger lobe (Fig. 109 cf. Fig. 117), by the longer palpal tibia (Fig. 1 12 cf. Fig. 116), and by the larger diameter of the embolic coil (Fig. 110 cf. Fig. 115). S. montanus male differs from S. sanctus only in the larger diameter of the embolic coil (Figs. 108, 110 cf. Figs. 118, 119). The female of S. sanctus has the epigynum generally similar in form to those of S. kenus , S. castoris, S. bicavatus, S. crinitus, S. boreus, S. bipoculatus, S. sintalutus, S. montanus, S. dubiosus and S. apache ; the separation of these species is based on small differences in the epigyna. In S. kenus and S. castoris the posterior plate is longitudinally rather shorter than in S. sanctus , with the cusps closer together in S. kenus and larger in S. castoris (Fig. 123 cf. Figs. 95, 122). S. bicavatus (Fig. 125) also has the posterior plate somewhat shorter longitudinally than in S. sanctus , while the cusps are larger and closer together. S. crinitus (Fig. 107) has the posterior plate very similar to S. sanctus , but the cusps are minute and widely separated; S. boreus (Fig. 127) is also very similar to S. sanctus , but the epigynum is significantly larger (width of plate ca. 0.36-0.42 mm, cf. 0.25-0.30 mm for S. sanctus and S. crinitus). The epigyna of S. sanctus and S. bipoculatus (Fig. 126) differ only by the much larger cusps present in the latter species. In S. sintalutus (Fig. 1 28) the markings on the posterior plate are somewhat wavy as in S. sacer Figs. 133-1 38. -Internal genitalia, females, ventral: 133, S. ambiguus’, 134, S. kenus’, 135, S. sanctus’, 136, S. bicavatus ; 137, S. formicarius; 138, S. boreus. (Scale lines 0.1 mm). 200 THE JOURNAL OF ARACHNOLOGY (Fig. 27), rather than smoothly curved as in S. sanctus (Fig. 123) and the cusps are slightly further apart. The epigynum of S. montanus is very close to that of S. sanctus ; the cusps are rather closer together (Fig. 123, cf. Fig. 124), but whether this is a constant difference is uncertain. In S. dubiosus (Fig. 129) the cusps are smaller and closer together than in S. sanctus , and the outlines of the posterior plate and of the median septum are less clearly defined. In S. apache (Fig. 130) the median septum is broader than in S. sanctus , and the markings on the plate are more wavy than in S. sanctus. Distribution.— This species has a wide distribution in the western half of N. America, and is one of the commoner species. It is recorded from British Columbia, Alberta, Oregon, Montana, Utah, Colorado, Wyoming and Arizona (Map 4). The records from British Columbia, Alberta and Oregon are based on females only, and need confirmation by the capture of males. Natural History.— The female of this species has been taken much more frequently than the male. Males have been found in August, September and October, females in March-October; the main season of maturity seems to be in summer and autumn. The species has been recorded at both high and low altitudes, in a number of habitats: at the edge of snow at 2500-3000 m, under stones by the shore of a lake, in woodland. Scotinotylus crinitus, new species Figures 104, 105, 106, 107; Map 5 Type.— Holotype male from Red Feather Camp, 8200 ft., Colorado, September 21, 1946 (C. C. Hoff); deposited in AMNH. Description.— The male and female were taken together. Total length: female 2. 1-2.2 mm, male 2.0 mm. Carapace: length: female 0.90 mm, male 0.95 mm. Orange. Male carapace raised into lobe (Figs. 104, 106) which is almost identical with that of S. sanctus , but the hairs on the anterior face of the lobe are longer. Abdomen: grey; epigastric plates with weak striae in female, and with clear, rather widely spaced striae in male. Sternum: orange-yellow with dusky margins. Legs: orange-brown. Tibial spines: female 2221, male 0221 but weak. Tml: female 0.38-0.40, male 0.35. Male palp: gen- erally similar to that of S. sanctus , but the tibia is longer (Fig. 105), and the embolic coil is fairly wide distally, as in S. montanus (Fig. 119). Female palp: tibia with 3 tricho- bothria, one very small. Epigynum: Fig. 107; the cusps are tiny and widely separated. Diagnosis.— S. crinitus is closely related to S. sanctus , and its diagnosis is dealt with under that species. Distribution.— Known only from the type locality (Map 5). Natural History. -Both sexes were taken in September, at an altitude of about 2500 m; nothing was recorded on habitat. Scotinotylus montanus , new species Figures 118, 119, 120, 124; Map 4 Type.— Male holotype from Tioga Pass, California, 10,000 ft. altitude, September 22, 1961 (W. J. Gertsch and W. Ivie); deposited in AMNH. Description. -Both sexes were taken together. Total length: female 2.0 mm, male 1.70-1.90 mm. Carapace: length: female 0.90-0.95 mm, male 0.82-0.90 mm. Orange- yellow to orange, with faint dusky markings. Male carapace raised into large lobe similar MILLIDGE-GENUS SCOTINOTYLUS 201 to that of S. sanctus. Abdomen: grey; epigastric plates with very weak striae in female, with clear striae in male. Sternum: yellow, with grey reticulations and margins. Legs: orange-yellow. Tibial spines: female/male 2221, but spines on male tibiae I and II very short. Tml: female 0.37-0.45, male 0.41-0.43. Male palp: Figs. 118, 119, 120; the tibia has one stout spine. Female palp: tibia with 2 trichobothria. Epigynum: Fig. 124. Diagnosis.— S. montanus is closely related to S. sanctus , and its diagnosis is dealt with under that species. Distribution.— Known only from two localities in California (Map 4) Natural History.— This species has been taken only at high altitudes (3000 m and above), and has been found under rocks. The male was taken in August and September, the female in September. Scotinotylus humilis , new species Figures 115, 116, 117; Map 4 Type.— Holotype male from Crater Lake National Park (?), Oregon, 1951 (Lowrie); deposited in AMNH. Description.— The male type, the only specimen known, is in rather poor condition, with one palp and several limbs missing; the remaining palp is slightly expanded. Total length: male 1.80 mm. Carapace: length: male 0.85 mm. Deep brown, with dusky ."’V' V-'V "if 144 Figs. 139-144.-5. formicarius : 139, male palp, ectal; 140, male carapace, lateral; 141, male palp, meso-ventral; 142, male palpal tibia, dorsal; 143, epigynum; 144, epigynum, pale specimen. Abbreviations: M, membraneous part of suprategular apophysis; T, tegulum. (Scale lines 0.1 mm). 202 THE JOURNAL OF ARACHNOLOGY markings; raised into a rather shallow lobe (Fig. 117). In the type specimen, the anterior median eyes are poorly developed. Abdomen: grey; epigastric plates with weak, closely spaced striae. Sternum: deep brown. Legs: brown. Tibial spines: male 2221 but weak. Tml: male 0.37. Male palp: the embolic coil is rather small in diameter (Fig. 115), and the tibia is relatively short (Fig. 116). Diagnosis. -This species is closely related to S. sanctus, and its diagnosis is dealt with under that species. Distribution.— Known only from the type locality (Map 4). Natural History.— Nothing recorded. Scotinotylus bicavatus, new species Figures 125, 136; Map 4 Type.-Holotype female from Paradise, Rainier National Park, Washington, September 12, 1965 (J. and W. Ivie); deposited in AMNH. Figs. 145-154.-145, 5. bicornis, male carapace, lateral; 146, 5. petulcus, male carapace, lateral; 147, 148, 5. monoceros , male carapace, lateral; 149, 5. monoceros, female carapace, dorsal; 150, 5. monoceros, female carapace, lateral; 151, S. monoceros, epigynum; 152,5'. bicornis, female carapace, dorsal; 153, 5. bicornis, female carapace, lateral; 154,5. monoceros, internal genitalia, female, ventral. (Scale lines 0.2 mm, except 151, 154, 0.1 mm). MILLIDGE-GENUS SCOTINOTYLUS 203 Description.— Only the female is known. Total length: female 2.75 mm. Carapace: length: female 1.15 mm. Orange-brown, with dusky markings and margins. Abdomen: grey; epigastric plates with weak striae. Sternum: orange-brown, suffused with grey. Legs: orange-brown. Tibial spines: female 2221. Tml: female 0.40. Female palp: tibia with 3 trichobothria, one minute. Epigynum: Fig. 125; the cusps are large and fairly close together. Internal genitalia: Fig. 136. Diagnosis.— S. bicavatus is closely related to S. sanctus, and its diagnosis is dealt with under that species. Distribution.— Known only from the type locality (Map 4). Natural History.— The female was taken in September; nothing was recorded on habitat. Scotinotylus bipoculatus , new species Figure 126; Map 4 Type.— Female holotype from Timberline Lodge, Mt. Hood, Oregon, September 16, 1949 (V. Roth); deposited in AMNH. Description.— Only the female is known. Total length: female 2.55-2.70 mm. Carapace: length: female 1.15-1.20 mm. Orange, with dusky markings and margins. Ab- domen: whitish grey; epigastric plates with weak, fairly closely spaced striae. Sternum: orange, with dusky margins. Legs: orange. Tibial spines: female 2221. Tml: female 0.45-0.47. Female palp: tibia with 3 trichobothria, one very small. Epigynum: Fig. 126; the cusps are fairly large. Diagnosis. -5. bipoculatus is closely related to S. sanctus , and its diagnosis is dealt with under that species. Distribution.— Known only from the type locality (Map 4). Natural History.— The female was taken in September; nothing was recorded on habitat. Scotinotylus boreus, new species Figures 127, 138; Map 5 Type.— Female holotype from Whitemouth River, near Hadashville, Manitoba, May 10, 1966 (G. A. Bradley); deposited in CNC, Ottawa. Description.— Only the female is known. Total length: female 3.0 mm. Carapace: length: female 1.25 mm. Orange-brown, somewhat darkened anteriorly. Abdomen: grey- black; epigastric plates with fairly clear, closely spaced striae. Sternum: orange-yellow, with dusky margins. Legs: orange. Tibial spines: missing. Tml: female 0.44-0.45. Female palp: tibia with 3 trichobothria. Epigynum: Fig. 127; the cusps are very small and widely separated. Internal genitalia: Fig. 138. Diagnosis.— S. boreus is closely related to S. sanctus , and its diagnosis is dealt with under that species. Distribution.— This species is known only from Manitoba and Alberta (Map 5). Natural History.— The female has been taken in May and June. Habitats recorded were inside an anthill, inside the nest of Formica obscuripes, and in grassland (in a pitfall trap). 204 THE JOURNAL OF ARACHNOLOGY Scotinotylus sintalutus, new species Figure 128; Map 5 Type.— Female holotype from Sintaluta, Saskatchewan, in pitfall trap. May 11 - June 13, 1963 (A. L. Turnbull); deposited in CNC, Ottawa. Description.— Only the female is known. Total length: female 2.1 mm. Carapace: length: female 0.90 mm. Brown, with faint dusky markings. Abdomen: grey; epigastric plates smooth. Sternum: pale yellow with dusky markings. Legs: yellow-brown. Tibial spines: female 2221. Tml: female 0.40. Female palp: tibia with 3 trichobothria, one very small. Epigynum: Fig. 128. Diagnosis.— S', sintalutus is closely related to S. sanctus , and its diagnosis is dealt with under that species. Figs. 155-158.-155, S. monoceros, male palp, ectal; 156, S. bicornis, male palp, ectal; 157, S. monoceros, male palp, meso-ventral; 158, S. bicornis, male palp, meso-ventral. Abbreviations: M, membraneous part of suprategular apophysis; T, tegulum. (Scale lines 0.1 mm). MILLIDGE-GENUS SCOTINOTYLUS 205 Distribution.— Known only from the type locality (Map 5). Natural History. -The female was taken in May June, in a pitfall trap in grassland. Scotinotylus dubiosus, new species Figure 129; Map 4 Type .-Female holotype from Logan, Utah, April 30, 1948 (G. F. Knowlton); deposit- ed in AMNH. Description.-Only the female is known. Total length: female 2.0 mm. Carapace: length: female 0.90 mm. Orange-brown, with dusky markings and margins. Abdomen: grey-black; epigastric plates smooth. Sternum: orange-brown, with blackish margins. Legs: orange-brown. Tibial spines: mostly missing. Tml: female 0.37. Female palp: tibia with 3 trichobothria, one small. Epigynum: Fig. 129. Diagnosis.— £. dubiosus is diagnosed by the epigynum, which is of the same general form as that of S. sanctus ; see S. sanctus diagnosis. Distribution.— Known only from the type locality (Map 4). Natural History.— The only female was taken in April; nothing was recorded on habitat. Scotinotylus apache (Chamberlin), new combination Figure 130 Disembolus apache Chamberlin 1948:527 (Fig. 162, not Fig. 163) Type .—Female holotype from (?) Willow Creek, near Burchill’s Ranch, Arizona, October 1928; in AMNH, examined. Description.— This species is known only from the holotype, which is in very poor condition. All that remains is a legless carapace and some fragments of abdomen, but fortunately the epigynum is still present. The description given here is therefore very incomplete. Total length: 2 mm (according to Chamberlin). Carapace: length: female 0.70 mm. Brown, with dusky markings. Female palp: tibia with 3 trichobothria. Epigynum: Fig. 130; the median septum is rather broad, and the markings on the plate are slightly wavy. Clearing of the epigynum shows that the spermathecal ducts follow more or less the same course as in S. sanctus. Diagnosis.— S. apache is closely related to S. sanctus , and its diagnosis is dealt with under that species. Distribution.— Known only from the type locality; it has not been possible to identify this locality, and hence no map is given for this species. Natural History.— The female was taken in October; nothing was recorded on habitat. Scotinotylus formicarius (Dondale and Redner), new combination Figures 7, 15, 137, 139, 140, 141, 142, 143, 144; Map 5 Cochlembolus formicarius Dondale and Redner 1972:1644 Type. -Male holotype from Bellingham, Washington, May 12, 1971 (G. D. Alpert);in AMNH, examined. 206 THE JOURNAL OF ARACHNOLOGY Description. -Total length: female 2. 5-3.0 mm, male 2.0-2.35 mm. Carapace: length: female/male 1.1 mm. Orange-brown to brown. Male carapace raised into shallow lobe (Fig. 140), with the clypeus projecting. Abdomen: grey-black; epigastric plates weakly striated in female, more strongly so in male. Sternum: orange-yellow, reticulated with grey. Legs: orange to orange-yellow. Tibial spines: female/male 1111, but very short on tibiae I in male. Tml: female 0.47, male 0.43-0.46. Male palp: Figs. 139, 141, 142; the tibia has one stout spine. Female palp: tibia with 2 trichobothria. Epigynum: Figs. 143, 144; rather pale-colored in some specimens. The internal ducts follow a sinuous course (Fig. 137). Diagnosis.-The male of S. formicarius is diagnosed by the presence of a single stout spine on the palpal tibia, coupled with the form of the palpal tibia (Fig. 142), of the palp (Fig. 139) and of the carapace (Fig. 140). The female differs from all other species in the genus by the tibial spinal formula (1 1 1 1 , cf. 2221 for the other species), and the identity is confirmed by the epigynum (Figs. 143, 144). Distribution.— This species is known from a number of localities in Washington (Dondale and Redner 1972), and from Wyoming (Map 5). Natural History.— Males and females were taken in March, April, May and September. Its habitat seems to be the nests of ants {Formica obscuripes Forel), though this habitat was not mentioned for the Wyoming specimens. Map 1.- Distributions of S. sacer, S. alienus and S. protervus in North America MILLIDGE-GENUS SCOTINOTYLUS 207 Scotinotylus bodenburgi (Chamberlin and Ivie), new combination Figure 131 ; Map 5 “Erigone” bodenburgi Chamberlin and Ivie 1947:38 Type.— Holotype female from Matanuska, Bodenburg Butte, Alaska, June 2, 1945 (J. C. Chamberlin); in AMNH, examined. Description.— Only the female is known. Total length: female 1.70 mm. Carapace: length: female 0.70 mm. Orange, with dusky markings and margins. Abdomen: grey; epigastric plates smooth. Sternum: orange-yellow, suffused with grey. Legs: orange- brown. Tibial spines: female 2221. Tml: female 0.40. Female palp: tibia with 3 tricho- bothria. Epigynum: Fig. 131; very pale in color, with small, widely spaced cusps which point forwards like arrowheads. Diagnosis.— The female of S. bodenburgi is grouped with S. sagittatus by the form of the epigynum, which has 2 forward-directed cusps (Fig. 131). These two species have very similar epigyna, which are however separable: the cusps of S. bodenburgi are both smaller and further apart than those of S. sagittatus (Fig. 132). In addition, the palpal tibia of S. bodenburgi has 3 trichobothria, while that of S. sagittatus has only 2; this difference may not however be constant. Distribution.— Known only from the type locality (Map 5). Natural History. -The female was taken in June; nothing was recorded on habitat. Map 2.- Distributions of S. alpinus, S. exsectoides and S. vernalis in North America 208 THE JOURNAL OF ARACHNOLOGY Scotinotylus sagittatus , new species Figure 132; Map 5 Type.— Female holotype from Mt. Washburn (north of summit), Wyoming, August 13, 1940 (W. Ivie); deposited in AMNH. Description.— Only the female is known. Total length: female 1.80 mm. Carapace: length: female 0.80 mm. Orange-brown, with dusky markings and margins. Abdomen: grey ; epigastric plates with very weak striae. Sternum: orange-yellow, with dusky margins. Legs: yellow-brown. Tibial spines: female 2221. Tml: female 0.40. Female palp: tibia with 2 trichobothria. Epigynum: Fig. 132; pale in color, with cusps pointing forwards like arrowheads. Diagnosis.— S. sagittatus has the epigynum similar to that of S. bodenburgi , and its diagnosis is dealt with under that species. Distribution.— Known only from the type locality (Map 5). Natural History.— The female was taken in August, at a high altitude (ca. 3150 m); the habitat was not recorded. Scotinotylus monoceros (Simon), new combination Figures 9, 147, 148, 149, 150, 151, 154, 155, 157; Map 5 Delorrhipis monoceros Simon 1884:697 Erigone monoceros : Keyserling 1886:156 Coreorgonal monoceros : Bishop and Crosby 1935:219; Roewer 1942:621; Bonnet 1956:1203 Cheraira willapa Chamberlin 1948:518 NEW SYNONYMY. The type of this species cannot be found, but there are 2 vials (unnamed as to species) in AMNH which correspond in locality and date with those given under “Other records” (Chamberlin 1948:519): the specimens in these vials are S. monoceros females. Chamberlin’s figure 1 16 also corresponds well with S. monoceros. Type.— Type locality, Washington Territory. Description.— Total length: female/male 2. 5-2. 6 mm. Carapace: length: female 1.2-1. 3 mm, male 1.35-1.55 mm (including the horn). Orange, with faint dusky markings. The ocular area of the female does not project significantly over the clypeus (Figs. 149, 150). The male carapace is elevated anteriorly, but there is no actual lobe; a “horn” projects from the clypeus, at an angle which is somewhat variable (Figs. 147, 148). Abdomen: grey to black; epigastric plates smooth. Sternum: orange, reticulated with black, and with black margins. Legs: orange-brown. Tibial spines: female 2221, male 0011. Tml: female 0.65-0.70, male 0.58-0.63. Male palp: Figs. 155, 157. The tibia bears one very stout spine, and the tibial apophysis is long and hooked distally. The patella is long and swollen distally on the ventral side, and there is a white membraneous extrusion meso-dorsally between patella and tibia. Female palp: tibia with 3 trichobothria, one small. Epigynum: Fig. 151 ; internal genitalia Fig. 154. Diagnosis.— The male of S. monoceros can be diagnosed at once by the form of the carapace, which has a rod-like apophysis (the “horn”) projecting from the clypeus (Figs. 147, 148), coupled with the form of the palp (Fig. 155). The palp shows only minor differences from those of S. bicornis (Fig. 156) and S . petulcus. The female is diagnosed by the epigynum, which has neither cusps nor a process arising from the anterior margin (Fig. 151). S. ambiguus female falls into the same section of the key, but the epigyna of these two species are quite distinct (Fig. 151 cf. Fig. 121) and confusion is impossible. The epigynum of S. monoceros is structurally indistinguishable from the of S. bicornis MILLIDGE-GENUS SCOTINOTYLUS 209 (and possibly also from that of S. petulcus when this is discovered). S. monoceros and S. bicornis females may however be separable by the form of the carapace: the ocular area projects over the clypeus to a greater extent in S. bicornis than in S. monoceros (Figs. 149, 150 cf. Figs. 152, 153); more specimens of S. bicornis female (taken in company with the male) are required to establish whether this difference is reliable. Map 3. -Distributions of S. pallidus, S . ambiguus, S. eutypus, S. patellatus, S. sacratus, S. gracilis, S. majesticus, S. regalis and S. magnificus in North America 210 THE JOURNAL OF ARACHNOLOGY Distribution.— This appears to be a relatively common species in the western coastal region; there are records from British Columbia, Washington, Oregon and California, but also two records from Idaho (Map 5). Natural History. -The males appear to be much less frequent than the females, and presumably have a shorter season of maturity. Males have been taken in January-April and August-November, and females in every month of the year. The only habitats men- tioned are in a seepage area beside a creek, in a cave, and in fir needles. Scotinotylus bicornis (Emerton), new combination Figures 145, 152, 153, 156, 158; Map 5 Delorrhipis bicornis Emerton 1923:242 Coreorgonal bicornis: Bishop and Crosby 1935:218; Roewer 1942:621; Bonnet 1956:1203 Type.— Male type from Terrace, British Columbia, October 1923 (Mrs. Hippisley); in MCZ, examined. Description.— Total length: female 2.70-3.0 mm, male 2.8 mm. Carapace: length: female 1.1-1.25 mm, male 1.45 mm. Orange to chestnut-brown, with dusky markings. In the female, the ocular area protrudes slightly over the clypeus (Figs. 152, 153). The male carapace (Fig. 145) has 2 distinct lobes, one carrying the anterior median eyes, the other arising from the clypeal region; both lobes bear numerous short bristly hairs. Abdomen: grey to black; epigastric plates smooth. Sternum: orange, reticulated with black, and with black margins. Legs: orange-brown. Tibial spines: female 2221, male 0111, but spine on tibia II very short. Tml: female 0.52-0.60, male 0.55-0.57. Male palp: Figs. 156, 158. The tibia bears one very stout spine, and the tibial apophysis is long and hooked distally. The patella is long and swollen distally on the ventral side, and there is a white mem- Map 4. -Distributions of S. pollucis, S. castoris, S. kenus, S. sanctus, S. humilis, S. montanus, S. bicavatus, S. bipoculatus and S. dubiosus in North America MILLIDGE-GENU S SCOTINOTYLUS 211 braneous extrusion between patella and tibia. Female palp: tibia with 3 trichobothria, one small. Epigynum: not distinguishable from that of S. monoceros. Diagnosis. -The male of S. bicornis can be diagnosed readily by the form of the carapace (Fig. 145), coupled with the form of the palp (Fig. 156). The female of S. bicornis cannot be distinguished from S. monoceros by the epigynum, but may be separ- able by the form of the carapace (see S. monoceros diagnosis). Distribution.-This species is rare in comparison S. monoceros , the only records being from the type locality, at an altitude of ca. 1400 m (Map 5). Natural History.— This species will probably prove to have a more northerly distribu- tion than S. monoceros. The males were taken in October, the females in June and October. The only habitat given (for a single female) is on moss on a rockslide. Scotinotylus petulcus, new species Figure 146; Map 5 Type.— Male holotype from Denny Creek, Snoqualine Pass, Washington, September 16, 1935 (Chamberlin and Ivie); deposited in AMNH. Map 5. -Distributions of S. formicarius, S. crinitus, S. boreus, S. monoceros, S. bicornis, S. petulcus, S. sagittatus, S. sintalutus and S. bodenburgi in North America 212 THE JOURNAL OF ARACHNOLOGY Description. -Only the male is known. Total length: male 2.55 mm. Carapace: length: male 1.42 mm. Orange-brown, raised anteriorly into a lobe which carries the posterior and anterior median eyes, and with a large, almost spherical lobe arising from the clypeus (Fig. 146); both lobes are furnished with numerous short bristly hairs. Abdomen: black; epigastric plates smooth. Sternum: orange, suffused with black. Legs: orange. Tibial spines: male 0011, very short. Tml: male 0.63-0.67. Male palp: not distinguishable from that of S. monoceros. Diagnosis.— The male of S. petulcus can be diagnosed at once by the unique form of the carapace, with the large bulbous projection from the clypeus (Fig. 146). The palp is practically identical in form to those of S. monoceros and S. bicornis. The female is unknown, but is likely to be closely similar to the females of S. monoceros and S. bicornis. Distribution. -Known only from the type locality (Map 5). Natural History.— Two males were taken in September; nothing was recorded on habitat. ACKNOWLEDGEMENTS I am indebted to the following colleagues for the loan of the material studied: Dr. N. I. Platnick, American Museum of Natural History, New York; Prof. H. W. Levi, Museum of Comparative Zoology, Harvard University; Dr. C. D. Dondale, Agriculture Canada, Bio- systematics Research Institute, Ottawa. Dr. Platnick and Dr. Dondale gave some essential help with the mapping, and Mr. F. R. Wanless, Mr. P. Hillyard (British Museum, Natural History) and Mr. G. H. Locket (Stock- bridge) provided me with some of the literature. LITERATURE CITED Banks, N. 1896. A few new spiders. Canadian Ent., 28:62-65 Banks, N. 1910. Catalogue of nearctic spiders. Bull. United States Nat. Mus., 72:1-80 Bishop, S. C. and C. R. Crosby. 1935. Studies in American spiders: miscellaneous genera of Erigoneae, I. J. New York Ent. Soc., 43:217-281 Bishop, S. C. and C. R. Crosby. 1938. Studies in American spiders: miscellaneous genera of Erigoneae, II. J. New York Ent. Soc., 46:55-107 Bonnet, P. 1955. Bibliographia araneorum. Toulouse, Vol. 2( 1 ) : 1-9 1 7 Bonnet, P. 1956. Bibliographia araneorum. Toulouse, Vol. 2(2) :9 1 8-1925 Bonnet, P. 1958. Bibliographia araneorum. Toulouse, Vol. 2(4):3027-4230 Braendegaard, J. 1937. Spiders (Araneina) from southeast Greenland. Medd. Gronland, 108(4): 1-15 Braendegaard, J. 1946. The spiders (Araneina) of east Greenland. Meed. Gronland, 121(15): 1-128 Bragg, P. D. and R. E. Leech. 1972. Additional records of spiders (Araneida) and harvestmen (Phalangida) for British Columbia. J. Ent. Soc. British Columbia, 69:67-71 Cambridge, O. P.-. 1894. On some new and rare Scotch spiders. Ann. Scottish Nat. Hist., 1894:18-25 Chamberlin, R. V. 1948. On some American spiders of the family Erigonidae. Ann. Ent. Soc. America, 41:483-562 Chamberlin, R. V. and W. Ivie. 1933. Spiders of the Raft River Mountains of Utah. Bull. Univ. Utah (Biol.), 23(4): 1-79 Chamberlin, R. V. and W. Ivie. 1945. Erigonid spiders of the genera Spirembolus, Disembolus and Bactroceps. Trans. Connecticut Acad. Arts Sci., 36:215-235 Chamberlin, R. V. and W. Ivie. 1947. The spiders of Alaska. Bull. Univ. Utah (Biol.), 37(10): 1-103 Crosby, C. R. 1925. in R. V. Chamberlin, New North American spiders. Proc. California Acad. Sci., (4)14:105-142 MILLIDGE-GENUS SCOTINOTYLUS 213 Crosby, C. R. 1929. Studies in North American spiders: the genus Cochlembolus (Araneina). Ent. News, 40:79-83 Crosby, C. R. and S. C. Bishop. 1933. American spiders: Erigoneae, males with cephalic pits. Ann. Ent. Soc. America, 26:105-182 Dondale, C. D. and J. H. Redner. 1972. A synonym proposed in Perimones, a synonym rejected in Walckenaera, and a new species described in Cochlembolus (Araneida: Erigonidae). Canadian Ent., 104:1643-1647 Emerton, J. H. 1882. New England spiders of the family Theridiidae. Trans. Connecticut Acad. Arts Sci., 6:1-86 Emerton, J. H. 1909. Supplement to the New England spiders. Trans. Connecticut Acad. Arts Sci., 14:171-236 Emerton, J. H. 1915. Canadian spiders, II. Trans. Connecticut Acad. Arts Sci., 20:145-160 Emerton, J. H. 1917. New spiders from Canada and the adjoining states. Canadian Ent., 49:261-272 Emerton, J. H. 1923. New spiders from Canada and the adjoining states, No. 3. Canadian Ent., 55:238-243 Holm, A. 1950. Studien iiber die Spinnenfauna des Tornetraskgebietes. Zool. bidrag. Uppsala, 29:103-213 Holm, A. 1958. Spiders (Araneae) from Greenland. Ark. Zool., Ser. 2, 1 1 (3 1 ) :5 25 -5 34 Holm, A. 1960. On a collection of spiders from Alaska. Zool. bidrag. Uppsala, 33:109-134 Holm, A. 1967. Spiders (Araneae) from West Greenland. Medd. Gronland, 184(1): 1-99 Holm, A. 1970. Notes on spiders collected by the “Vega” expedition 1878-1880. Ent. scandinav., 1:188-208. Hull, J. E. 1911. New and rare British spiders. Trans. Nat. Hist. Soc. Northumberland (N. S.), 4(1 9 14): 42-5 8 (separata were issued in 1911) Ivie, W. 1965. Some synonyms in American spiders. J. New York Ent. Soc., 75(3) : 126-1 3 1 Jackson, A. R. 1933. Results of the Oxford University expedition to Akpatok in 1931. Araneae. Proc. zool. Soc. London, 1933(1) : 145-159 Keyserling, E. 1886. Die Spinnen Amerikas, II. Theridiidae, Part 2 Niirnberg. pp. 1-295 Koch, L. 1879. Arachniden aus Siberien und Novaja Semlja, eingesammelt von der schwedischen Expedition im Jahre 1875. Kongl. Svenska Vet.-Akad. Handl., 16(5 ): 3-1 36 Kulczynski, W. 1885. Araneae in Camtschadalia a Dre B. Dybowski collectae. Pam. Akad. umiej Krakow, 11:1-60 Marx, G. 1890. Catalogue of the described Araneae of temperate North America. Proc. United States Nat. Mus., 12:497-594 Merrett, P. 1963. The palpus of male spiders of the family Linyphiidae. Proc. zool. Soc. London, 140(3): 347-467 Millidge, A. F. 1977a. The conformation of the male palpal organs of Linyphiid spiders, and its application to the taxonomic and phylogenetic analysis of the family (Araneae: Linyphiidae). Bull. British arach. Soc., 4(1): 1-60 Millidge, A. F. 1977b. The genera Mecopisthes Simon and Hypsocephalus n. gen. and their phylo- genetic relationships (Araneae: Linyphiidae). Bull. British arach. Soc., 4(3) : 1 13-123 Millidge, A. F. 1980. The erigonine spiders of North America. Part 2. The genus Spirembolus Cham- berlin (Araneae: Linyphiidae). J. Arachnol., 8:109-158 Roewer, C. R. 1942. Katalog der Araneae. Bremen. Vol. 1, 1040 pp. Simon, E. 1884. Les Arachnides de France. Paris. Vol. 5, part 3:421-885 Soerensen, W. 1898. Arachnida Groenlandica (Acaris exceptis). Vid. Medd. naturh. Foren. Kjobenhavn, 1898:176-235 Thaler, K. 1970. Uber einige wenig bekannte Zwergspinnen aus den Alpen (Arach., Araneae, Erigonidae). Ber. Nat.-Med. Ver. Innsbruck, 58:255-276 Vogelsanger, T. 1944. Beitrag zur Kenntnis der schweizerischen Spinnenfauna. Mitt. Naturf. Ges. Schaffhausen, 19:158-190 Manuscript received January 1980, revised April 1980. Hadley, N. F., G. A. Ahearn, and F. G. Howarth. 1981. Water and metabolic relations of cave-adapted and epigean lycosid spiders in Hawaii. J. Arachnol., 9:215-222. WATER AND METABOLIC RELATIONS OF CAVE-ADAPTED AND EPIGEAN LYCOSID SPIDERS IN HAWAII1 Neil F. Hadley Department of Zoology Arizona State University, Tempe, AZ 85281 Gregory A. Ahearn Department of Zoology University of Hawaii, Honolulu, HI 96822 and Francis G. Howarth Department of Entomology Bernice P. Bishop Museum, Honolulu, HI 96818 ABSTRACT Water loss rates, cuticular permeability, metabolic rates and rhythms were determined for the troglobitic spider, Lycosa howarthi, and an undescribed epigean spider, Lycosa sp., collected from lava tube caves and lava flows, respectively, on the Island of Hawaii. The cuticular lipid and hydrocarbon surface densities as well as an analysis of the hydrocarbon fraction were also determined for each species and correlated with their different cuticular permeabilities. Water loss for the cave species was significantly higher at each test humidity, with the maximum difference between the species occuring at 19°C and 0% RH (11.14 ± 1.76 mgg-1 h_1 vs 1.01 ± 0.33 mg g_1 h_1 ). Water loss rates decreased with decreasing saturation deficit between 0 and 70% RH in the cave species, but were relatively independent of RH in the epigean species. The mean metabolic rate for a 12-h period between 1500 and 0300 h was approximately one and one-half times greater in the epigean species (173.72 ± 6.22 jul 02 g"1 h"1 vs 115.71 ± .89 jul 02 g_1 h'1 ). Oxygen consumption exhibited an increased trend in the epigean species during hours of darkness; 02 consumption rates for the cave species were very constant over the 12-h period. The surface densities of cuticular lipid and cuticular hydrocarbon were significantly greater in the epigean species on both an individual spider and weight-specific basis. The hydrocarbon fraction of both species was comprised of numerous components, all saturated, ranging in chain length from 18-19 to over 41 carbon atoms. Straight-chain («-alkane) molecules accounted for almost 65% of the total hydrocarbon fraction in the cave species, but only 23.5% of the total in the epigean species. Supported by NSF Grant PCM77-23803 to NFH and NSF Grant DEB79-04760 to FHG. 216 THE JOURNAL OF ARACHNOLOGY INTRODUCTION Most deep cave zones are characterized by perpetual darkness, stable temperatures, and constant high humidity. In view of these conditions, it has often been assumed that species restricted to these zones (troglobites) exhibit physiological responses that mirror the physical environment. Studies of cave insects and millipedes which show that they tend to select microenvironments that are high in relative humidity and relatively cool (Mitchell 1971, Bull and Mitchell 1972, Wilson 1975) add credence to this assumption. Experimental data, however, are still insufficient to permit firm conclusions regarding preference and tolerance capacities. The unique cave faunal community discovered in lava tubes on the Hawaiian Islands (Howarth 1972, 1973) not only provides an opportunity to examine thermal, water and metabolic relations of troglobitic arthropods, but also permits comparison with closely related, perhaps ancestral, epigean species which exist near lava caves. One such example involves wolf species (Lycosidae). Lycosa howarthi Gertsch, the largest terrestrial troglo- bite in lava tubes on Hawaii Island, exhibits loss of pigment and has vestigal, probably non-functional eyes. A second, presently undescribed lycosid spider ( Lycosa sp.) occurs on the surface of younger lava flows on Hawaii Island in some cases adjacent to the entrances of lava tubes inhabited by L. howarthi. This epigean species possesses the very large eyes typical of lycosids and is subjected to much greater fluctuations in temperature and humidity than its troglobitic congener. This paper contains data on water loss rates, cuticular permeability, and metabolic rates and rhythms in these two spider species. Cuticular lipid and hydrocarbon surface densities as well as an analysis of the hydrocarbon fraction were determined for each species and correlated with observed differences in cuticular permeability. METHODS Spiders were collected on the Island of Hawaii between 1 September and 1 November 1979. Specimens of Lycosa howarthi came from the deep cave zone of Kazumura Cave (elev. 300—400 m) where temperatures remain between 19 and 20°C. The specimens were all mature or nearly mature females. The epigean lycosid was collected at night on the largely unvegetated lava flows on Kilauea Volcano between 800 and 950 m within Hawaii Volcanoes National Park. The specimens of Lycosa sp. were mostly mature or nearly mature females. Cave spiders were maintained in the laboratory in individual plastic vials at 19°C and 100% RH; surface spiders were kept in individual vials at room temperatures (ca. 25°C) and humidities (ca. 50-60% RH). Water loss rates were determined gravimetrically on spiders placed inside a sealed glass desiccator maintained at 19°C and at relative humid- ities of approximately 0, 50, 70 and 90%, the latter generated by either Drierite (0%) or saturated salt solutions (Winston and Bates 1960). Relative humidity during each expo- sure was monitored by a standard laboratory hygrometer located inside the desiccator. A 5-hr incubation period was used in all tests. Respiration was measured at 19°C using Gilson respirometer flasks containing 0.5 ml of 10% KOH to absorb C02. The flasks were lined with water-soaked tissue paper to maintain a near-saturated atmosphere during the exposures. A black plastic sheet was placed over the respirometer during all experimental runs to exclude light. HADLEY, AHEARN AND HOWARTH-WATER AND METABOLIC RELATIONS 217 Table 1.- Lipid/hydrocarbon quantities extracted from the epicuticles of Lycosa hoxvarthi and Lycosa sp. Species N Pooled weight (g) Total lipid (mg) Hydro- carbon (HC) (mg) Lipid/ spider (mg) HC/ spider (mg) HC/ lipid (%) Lycosa sp. 12 1.3096 2.12 0.24 0.18 0.02 11.3 (epigean) L. hoxvarthi 9 1.7575 1.11 0.13 0.12 0.01 11.7 (cave) Epicuticular lipids were removed by slurrying the spiders (pooled) in redistilled hexane for two 10-min perods. The lipid extract was filtered, evaporated to dryness under nitro- gen, and weighed to 0.01 mg. Lipid classes present were determined by spotting a 1 pi sample of the extract on 0.25 mm Silica Gel G coated glass plates (TLC) and developing the plates in hexane: diethyl ether:formic acid (80:20:2, v/v). Lipid bands were detected by charring and identified against known standards. Hydrocarbons were separated from the other lipid classes by eluting the lipid extract with hexane through silicic acid col- umns (Jackson et al. 1974). The amount of hydrocarbon was determined gravimetrically after drying under nitrogen and an aliquant spotted on silver nitrate impregnated TLC plates to check for unsaturation. The hydrocarbon fraction was analyzed by gas chroma- tography (GLC) using 6 ' by 1/8" in. glass columns packed with 3% OV-101 on 100/120 Gas Chrom Q and programmed from 200 to 300°C at 4°C/min. Peaks were identified by comparison to retention times of standards and quantified by electronic integration. Branched components were identified on the basis of franctional equivalent chain lengths, as there was insufficient material to permit separation of ^-alkanes and branched alkanes using the standard molecular sieve technique (Hadley and Jackson 1977). RESULTS Water Loss.— The water loss rates (WLR) for Lycosa hoxvarthi (troglobitic) and Lycosa sp. (epigean) spiders are compared in Fig. 1 . WLR were significantly higher (P < 0.01) for the cave species at each humidity, with the maximum difference occurring at the highest saturation deficit (0% RH). At 19°C and 90% RH, four of six epigean spiders exhibited a slight gain in body mass. The net result was an apparent water gain for this test group, whereas all cave spiders under identical conditions continued to lose water. WLR for the cave species were also significantly higher at each humidity when rates were expressed per unit surface area. Species differences in area-specific WLR followed the pattern observed for weight-specific determinations. Oxygen Consumption. -Metabolic rates for the two spider species for a 12-h period between 1500 and 0300 hours are presented in Fig. 2. The mean weight (± 1 SD) for the troglobitic L. hoxvarthi was 184.4 ± 16.0 mg (n = 4) versus 123.0 ± 24.0 mg for the epigean Lycosa sp. (n = 4). Oxygen comsumption expressed per unit spider mass was significantly higher (P < 0.01) for the epigean spiders at each hour of measurement. The mean value (± 1 SE) for the 12-h period for epigean spiders was 173.72± 6.22 pi 02 g"1 h_1 compared to 1 15.71 ± .89 pi 02 g"1 h'1 for the troglobitic spiders. These data indicate a metabolic rate that is approximately 1.5 times greater in the surface-dwelling spiders. In 218 THE JOURNAL OF ARACHNOLOGY the latter group there also was a trend for increased metabolism in hours which corre- spond to darkness in their natural habitat. In contrast, oxygen consumption rates were very constant for the cave species over the entire 12-h period. Cuticular Lipids.— Chromatographic separation of extracted epicuticular lipids indi- cated the presence of hydrocarbons, wax esters, alcohol, sterols (principally cholesterol), triglycerides and free fatty acids in both spider species. Lipid-hydrocarbon quantities for the two species are given in Table 1. Epigean spiders (pooled) had 1.9 times more total li- pid and 1.8 times more total hydrocarbon than the cave spiders. On a per individual basis, an epigean spider had 1.5 times as much lipid and twice as much hydrocarbon as its cave counterpart. Expressed per unit body weight, these values convert to 1.93 mg g_1 vs. 0.63 mg g_1 (lipid) and 0.18 mg g_1 vs. 0.07 mg g_1 (hydrocarbon) for the epigean and cave Table 2. -Hydrocarbon composition (%) of Lycosa howarthi and Lycosa sp. cuticular lipids. Only components accounting for 0.1% or greater of the total HC fraction are listed. Values represent means of three replicate runs on each group. ECL = equivalent chain length; tr = trace (less than 0.1%);!b = branched. GLC peak no. ECL Lycosa howarthi Lycosa sp. 18 18 .40 tr 19 19 1.07 .22 19b 19.2 — .37 19b 19.6 .51 .54 20 20 3.13 .83 21 21 4 38 1.92 21b 21.6 2.98 1.45 22 22 3.17 1.37 22b 22.4 2.85 1.60 23 23 4.33 1.47 23b 23.3 — .83 23b 23.7 — .62 24 24 4.36 1.39 24b 24.3 1.72 .53 24b 24.7 — .65 25 25 4.05 2.22 25b 25.3 — .45 25b 25.7 .70 .56 26 26 2.84 1.26 26b 26.4 .40 — 26b 26.7 .54 1.57 27 27 3.29 4.01 27b 27.3 — .49 27b 27.7 — .95 28 28 4.24 1.99 28b 28.3 — .15 28b 28.7 .47 1.75 29 29 6.24 6.21 29b 29.3 — 3.86 29b 29.7 — 3.15 30 30 5.06 .60 30b 30.4 — .34 31 31 5.03 .77 31b 31.3 — 1.85 31b 31.6 — 1.86 32 32 3.67 .41 32b 32.3 — 2.18 32b 32.7 — .50 33 33 3.43 .75 33b 33.3 — 3.00 33b 33.7 .54 9.71 34 34 2.33 tr 34b 34.3 — 9.48 34b 34.7 .82 .66 35 35 2.20 — 35b 35.5 1.18 5.14 36 36 1.41 — 36b 36.4 1.85 2.08 37 b 37.3 1.93 2.81 37b 37.7 2.50 7.17 38b 38.4 4.70 .52 39b 39.5 3.80 5.71 40b 40.3 1.40 2.13 40b 40.8 1.01 — 41b 41.7 5.05 — n -alkanes 64.9% 23.5% branched alkanes 35.1% 76.5% HADLEY, AHEARN AND HOWARTH-WATER AND METABOLIC RELATIONS 219 species, respectively. The ratio of hydrocarbon to lipid was virtually identical for both species. No attempt was made to quantify or further analyze any of the non-hydrocarbon cuticular constituents. Gas chromatographic analysis of the hydrocarbon fraction of Lycosa sp. revealed 48 separable components versus 39 for Lycosa howarthi (Table 2). In both species the hydro- carbon molecules were saturated and ranged from approximately 18 to over 40 carbon atoms in length. A major difference was the predominance of straight-chain (/7-alkane) molecules in the cave species and branched molecules in the epigean species. This differ- ence resulted from the absence of branched components that correspond to the //-alkanes of the same carbon number (especially between C27 and C33) in the cave species (Table 2). Long-chain branched molecules characterized the hydrocarbon fraction of both species. The hydrocarbon components were present in relatively equal amounts, particu- larly in the cave species where only one molecule accounted for more than 6% of the total hydrocarbon fraction. Although there was not enough material to permit mass spectroscopic confirmation of branching types, fractional equivalent chain lengths (0. 3-0.4 and 0. 6-0.7) suggest the presence of at least two homologous series of methyl branched components. DISCUSSION The habitats occupied by the two lycosid spiders examined in this study represent highly contrasting physical environments. The true deep cave zone, inhabited by Lycosa howarthi , features a relatively stable climate where temperatures are moderate (19 ±1°C) and evaporation absent or negligible. It is a rigorous environment in that it is perpetually dark and food-limited. Ecological studies including a food web analysis were reported in Howarth (1973) and Gagne and Howarth (1975). The epigean wolf spider ( Lycosa sp.) occurs on barren to semi-vegetated flows of Kilauea Volcano (Howarth 1979). Although annual rainfall ranges between 1100 and 2000 mm, the surface appears extremely xeric due to rapid evaporation from the black lava, the poor water-holding capacity of lava, the rapid percolation of rain water into the porous rock, and the high drying power created by the almost constant wind. A 25° to 50°C difference between the daily maximum and minimum surface temperatures is not uncommon. It is not surprising that Lycosa sp. individuals hide in cracks and under large rocks during the day and only venture onto the surface at night. The water loss rates (WLR) for the two spider species correspond to the environmental stresses imposed by their respective habitats. In the troglobitic spiders, which normally experience saturated atmospheres, WLR increased markedly with a reduction in relative humidity (i.e., increased saturation deficit), reaching a maximum of 11.14 mgg-1 h_1 at 19°C and 0% RH (Fig. 1). In contrast, WLR for the lava flow spiders, which are subjected to higher saturation deficits, were relatively independent of changes in relative humidity between 0 and 70% (Fig. 1). Higher transpiration rates were also found in the cave- dwelling cockroach Blaberus craniifer Burmeister (Herreid 1969) and the troglobitic col- lembolan Tomocerus problematicus (Vannier 1979) than in their epigean counterparts Periplaneta americana (L.) and Tomocerus minor (Lubbock), respectively. Under the experimental conditions employed in this study (low temperature, post- absorptive animals), the total water loss rates for the spiders essentially represent their cuticular transpiration rates and, hence, permit discussion in terms of cuticular perme- ability. For this purpose it is convenient to use area-specific water loss rates corrected for 220 THE JOURNAL OF ARACHNOLOGY Fig. l.-Mean water loss (gain) rates of the troglobitic spider, Lycosa howarthi, and the lava flow spider, Lycosa sp. at 19°C and various relative humidities. Sample size at each test humidity was seven except for 90% (n = 6). Vertical bars repre- sent ± 1 SE. % RELATIVE HUMIDITY saturation deficit (i.e., pg cm-2 h"1 mmHg-1 ) (Edney 1977). At 19°C and 0% RH, perme- ability values for the troglobitic spider were 33.4 pg cm-2 h_1 mmHg-1 versus 3.1 pg cm-2 h_1 mmHg"1 for the epigean lycosid. Although the cuticular permeability of the cave species is 10 times higher than the lava flow species, it is only slightly greater than the permeability for the mesic-adapted spider, Lycosa amentata (Clerck) (28.3 pg cm-2 h"1 mmHg-1 ) (Davies and Edney 1952), and is significantly lower than values given for several species of Australian cave-dwelling crickets (Campbell 1980). The cuticular permeability of the lava flow spider is typical of the most xeric-adapted insects and arachnids (Edney 1977) and no doubt is an important factor in their ability to inhabit the harsher surface lava environment. The difference in cuticular permeability between Lycosa howarthi and Lycosa sp. is likely due in part to differences in their surface lipid and hydrocarbon densities. The epigean species possessed greater amounts of both cuticular lipid and hydrocarbon on an individual spider and weight -specific basis (Table 1). The hydrocarbon fraction, which is effective in waterproofing the surface of many plants and animals (Hadley 1980), was nearly three times more abundant in the epigean spiders (weight-specific) and nearly twice as abundant when expressed per unit surface area (.007 vs .004 mg cm-2). The hydrocarbon density for Lycosa sp. is virtually identical to values for xeric-adapted black widow spiders, Latrodectus hespems Chamberlin and Ivie (Hadley 1978). The correlation between cuticular permeability and the composition of the hydrocarbon fractions of the two Hawaii Island Lycosa species is not as clearly defined. Hydrocarbons of both species were composed of numerous, saturated components, and the range of chain lengths represented were essentially the same (Table 2). A notable difference between the two spider species was the abundance of branched alkanes in the epigean spiders and the predominance of /7-alkanes in the cave spiders. A higher percentage of long-chain, branch- ed molecules characterized the hydrocarbon fraction of the desert-adapted scorpion, Hadrurus arizonensis, Ewing and another scorpion species, Centruroides sculpturatus Ewing, active during hot summer months when compared with a mesic-adapted scorpion, Uroctonus apacheanus Gertsch and Soleglad, and C. sculpturatus, collected during winter, HADLEY, AHEARN AND HOWARTH-WATER AND METABOLIC RELATIONS 221 respectively (Toolson and Hadley 1977, 1979). In the present study, however, the higher frequency of branched molecules in the epigean species is evident only in the mid-range of its hydrocarbon spectrum (Table 2). Compositional differences in the hydrocarbon fraction of the two spider species do indicate either major differences in their diet and/or genetic isolation over a sufficiently long period to permit changes in the enzymes respon- sible for synthesis of the branching types present. The metabolic rate data for the two spider species are very preliminary and must be viewed with caution even though observed rates coincide with published values for arach- nids and specifically spiders under comparable test conditions. The mean (±1 SE) oxygen consumption for the lava flow spider (173.72 ±6.22 ql 02 g_1 h_1 ) over the 12-h period from 1 500 to 0300 h (19°C) is significantly higher (P < 0.01) than the mean recorded for the cave species (115.71 ± .89 jid 02 g_1 h"1 ). Both values fall within the range of metabolic rates (21-356 jul 02 g_1 h_1) reported for arachnids at 20°C by Anderson (1970) as well as for lycosid spiders (92.8-452.2 /il 02 g_1 fr1 ) at 20°C (calculated from data presented by Humphreys 1977). The metabolic rates of arachnids are typically lower than those of poikilotherms of comparable size and, in the case of the cave species, falls at the lower end of the metabolic spectrum for lycosid spiders. This finding is consistent with the hypothesis (Anderson 1970) that a comparatively low metabolic rate for arach- nids (spiders) is an adaptation to the potential problem of being faced with an inconsist- ent food supply, a situation likely to occur in the deep cave zone of the Hawaiian lava tubes (Howarth 1973). It is also consistent with the “sit and wait” predatory behavior exhibited by Lycosa howarthi in the cave environment (Howarth, unpub.). Fig. 2. -Mean oxygen consumption rates of the troglobitic spider, Lycosa howarthi (closed circles) and the lava flow spider, Lycosa sp., (open circles) at 19°C between 1500 h (3 PM) and 0300 h (3AM). Vertical bars represent ± 1 SE. Minimum sample size for each species was four. 222 THE JOURNAL OF ARACHNOLOGY The preliminary metabolic data (Fig. 2) suggest an increased oxygen consumption rate for the epigean species at approximately the time it would be active in nature, and the apparent lack of any cyclic rhythm in the cave species. A cyclic rhythm in oxygen consumption could be expected in the epigean species which routinely leaves the protec- tion of daytime cover and moves to the surface at night to feed. The apparent absence of any metabolic rhythm in the cave species is also not surprising since it remains in a stable environment with no obvious environmental cue to initiate feeding or locomotor behav- ior. Additional tests using larger sample sizes, a variety of ambient conditions, and spiders segregated as to sex and age are necessary to confirm these preliminary observations. LITERATURE CITED Anderson, J. F. 1970. Metabolic rates of spiders. Comp. Biochem. Physiol., 33:51-72. Bull, E. and R. W. Mitchell. 1972. Temperature and relative humidity responses of two Texas cave- adapted millipedes, Cambala speobia (Cambalia: Cambalidae) and Speodesmus bicornourus (Poly- desmida: Vanhoeffeniidae). Int. J. Speleol., 4:365-393. Campbell, G. D. 1980. Aspects of the ecology, environmental physiology and behavior of several Australian cave-dwelling crickets. Ph.D. Dissertation. University of New South Wales, Kensington, Australia. Davies, M. E. and E. B. Edney. 1952. The evaporation of water from spiders. J. Exp. Biol. 29:571-582. Edney, E. B. 1977. Water Balance in Land Arthropods. Springer-Verlag, New York. Gagne, W. C. and F. G. Howarth. 1975. The cavernicolous fauna of Hawaiian lava tubes, 6. Meso- veliidae or water treaders (Heteroptera). Pacific Insects, 16(4): 399-41 3. Hadley, N. F. 1978. Cuticular permeability and lipid composition of the black widow spider, Latro- dectus hesperus, Pp. 429-438, In Symp. Zool. Soc. Lond., 42 (Arachnology), P. Merrett (ed.). Academic Press, New York. Hadley, N. F. 1980. Surface waxes and integumentary permeability. Am. Sci., 68:546-55 3. Hadley, N. F. and L. L. Jackson. 1977. Chemical composition of epicuticular lipids of the scorpion, Paruroctonus mesaensis. Insect Biochem., 8:17-22. Herreid, C. F., II. 1969. Water loss of crabs from different habitats. Comp. Biochem. Physiol., 28:829-839. Howarth, F. G. 1972. Cavernicoles in lava tubes on the Island of Hawaii. Science, 175:325-326. Howarth, F. G. 197 3. The cavernicolous fauna of Hawaiian lava tubes, 1. Introduction. Pacific Insects, 15(1): 1 39-151. Howarth, F. G. 1979. Neogeoaeolian habitats on new lava flows on Hawaii Island: an ecosystem supported by windborne debris. Pacific Insects, 20(2-3): 133-144. Humphreys, W. F. 1977. Respiration studies on Geolycosa godeffroyi (Araneae: Lycosidae) and their relationship to field estimates of metabolic heat loss. Comp. Biochem. Physiol., 57A:255-263. Jackson, L. L., M. T. Armold, and F. E. Regnier. 1974. Cuticular lipids of adult fleshflies Sarcophaga bullata. Insect Biochem., 4:369-379. Mitchell, R. W. 1971. Preference responses and tolerances of the troglobitic carabid beetle, Rhadine subterranea. Int. J. Speleol., 3:289-304. Toolson, E. C. and N. F. Hadley. 1977. Cuticular permeability and epicuticular lipid composition in two Arizona vejovid scorpions. Physiol. Zool., 50:323-330. Toolson, E. C. and N. F. Hadley. 1979. Seasonal effects on cuticular permeability and epicuticular lipid composition in Centruroides sculpturatus Ewing 1928 (Scorpiones: Buthidae). J. Comp. Physiol., 129:319-325. Vannier, G. 1977. Relations hydriques chez deux especes de Tomoceridae (Insectes Collemboles). Peuplant des niveaux ecologiques separes. Bull. Soc. Zool. Fr., 102:63-79. Wilson, J. 1975. The effect of low humidity on the distribution of Heteromurus nitidus (Collembola) in Radford Cave, Devon. Trans. Brit. Cave Res. Assoc., 2:123-126. Winston, P. W. and D. H. Bates. 1960. Saturated solutions for the control of humidity in biological research. Ecology, 41:232-237. Manuscript received May 1980, revised August 1980. LourenQO, W. R. and M. Vachon. 1981. Complements a la description & Acanthothraustes brasiliensis (Mello-Leitao 1931) (= Teuthraustes brasilianus Mello-Leitao 1931), synonyme d'Euscorpius flavi- caudis (Geer 1778) (Scorpiones, Chactidae). J. Arachnol., 9:223-228. COMPLEMENTS A LA DESCRIPTION ^ACANTHOTHRAUSTES BRASILIENSIS (MELLO-LEITAO 1931) (= TEUTHRAUSTES BRASILIANUS MELLO-LEITAO 1931), SYNONYME d 'EUSCORPIUS FLA VICA UDIS (GEER 1 778) (SCORPIONES, CHACTIDAE) Wilson R. Lourenqo1 Laboratoire de Zoologie, Ecole Normale Superieure 46, rue d’Ulm 75005 Paris, France and Max Vachon Laboratoire de Zoologie (Arthropodes) Museum National d’Histoire Naturelle 61, rue de Buffon 75005 Paris, France ABSTRACT The comparative study of the types of Acanthothraustes brasiliensis (Mello-Leitao 1931) and of Euscorpius flavicaudis (Geer 1778) demonstrates that the first one is a synonym of E. flavicaudis. Our conclusions are mostly taken from the trichobothrial pattern analysis and also from some other less important characters. RESUME La comparaison des caracteres, specialement ceux tires de la trichobothriotaxie, du type d' Acan- thothraustes brasiliensis (Mello-Leitao 1931), du type et de nombreuses specimens Euscorpius flavi- caudis (Geer 1778), prouve l’identite de ces deux especes, si inattendue soit-elle. La synonymie suivante est discutee et proposee: Acanthothraustes brasiliensis (Mello-Leitao 1931) ~ Euscorpius flavicaudis (Geer 1778). INTRODUCTION L’espece Teuthraustes brasilianus a ete decrite par Mello-Leitao (1931), apres etude d’un seul exemplaire male qui, selon lui, habitait l’Etat du Para. Aucune localite typique n’est precisee, aucun renseignement n’est donne sur les conditions de recolte de l’exem- plaire type. Cette espece etait la premiere du genre Teuthraustes decrite pour le Bresil. Attache au Laboratoire de Zoologie (Arthropodes) du Museum National d’Histoire Naturelle. 224 THE JOURNAL OF ARACHNOLOGY Dans son travail sur les Scorpions sud-americains, Mello-Leitao (1932) mentionne T. brasilianus dans une liste des especes de Chactidae. L’espece est, a nouveau citee dans sa monographic (1945) et a ce moment, Mello-Leitao propose de placer T. brasilianus dans un nouveau genre: Acanthothraustes dont le genotype serait l’espece brasilianus , nom qu’il modifie en brasiliensis. Ce changement de nom ne se justifie pas, meme lors de la creation d’un nouveau genre, vu que la racine des deux genres est la meme. Un detail a signaler est que l’etiquette originale du type porte le nom de brasiliensis et non celui de brasilianus. Pourquoi Mello-Leitao a choisi un nom different dans sa premiere description? Nous ne le savons pas. De toute fa£on, le terme correct nous semble etre brasilianus. Cette erreur de nomen- clature sera neanmoins sans importance, apres la conclusion a laquelle nous sommes arrives. En examinant les collections du Museu Nacional de Rio de Janeiro, l’un de nous (WRL) a retrouve le type de T. brasilianus , en tres mauvais etat malheureusement. L’examen de cet exemplaire, nous a profondement etonnes, puisqu’il nous a conduits a constater qu’il s’agissait en fait d’un Euscorpius flavicaudis, Scorpion tres commun dans la region mediterraneenne du sud de la France (Fig. 1). Fig. 1.- Euscorpius flavicaudis male (RS-3218), vue dorsale. LOURENCO AND VACHON -ACANTHOTHRAUSTES BRASILIENSIS SYNONYMIE 225 ETUDE COMPAREE DES CARACTERES RELEVES CHEZ LE TYPE d 'ACANTHOTHRAUSTES BRASILIENSIS (BRASILIANUS) ET LE TYPE dEUSCORPIUS FLA VIC A UDIS. Trichobothriotaxie — Nous donnons priorite a la trichobothriotaxie parce qu’elle est toujours identique chez les deux sexes d’une meme espece et parce qu’elle ne subit aucune modification au cours du developpement postembryonnaire (invariance ontogenetique). Afin d’arriver a une conclusion sure, nous avons compare la trichobothriotaxie d'Acan- thothraustes a celle des 31 genres connus actuellement de nous appartenant a la famille des Chactidae mais aussi a celle des Vaejovidae, Tun de nous (MV) ayant depuis de nombreuses annees souligne l’absence de veritables caracteres permettant de separer ces deux families avec une absolue certitude. La premiere conclusion a laquelle nous fait aboutir la comparaison des trichobothrio- taxie s est que le genre Acanthothraustes Mello-Leitao 1945 est synonyme du genre Euscorpius Thorell 1876. Les preuves suivantes peuvent etre donnees: Chez tous les Euscorpius , la face externe de Tavant-bras (tibia) possede six territoires trichobothriotaxiques alors que chez tous les autres genres, il n’en existe que cinq (sauf bien entendu chez les genres ou le nombre tres eleve de trichobothries ne permet pas leur repartition en territoires; Dasyscorpiops et Hadrurus, par exemple). Le territoire supple- mentaire a ete defini par Tun de nous (Vachon 1973) et porte le sigle eba‘, il est situe entre le territoire eb (Fig. 15) et le territoire esb , lequel se reconnaft toujours par la possession d’une microtrichobothrie esb2 . Il suffit de comparer les figures 14 et 15 pour constater que chez Acanthothraustes le territoire accessoire eb a est present (Vachon 1975). Une autre preuve de la synonymie de ces deux genres est fournie par l’emigration de certaines trichobothries de la face ventrale de la main sur la face externe de ladite main. Et ceci est un caractere propre aux Euscorpius. La figure 7 relative a Euscorpius flavi- caudis montre l’existence d’une serie lineaire de trichobothries ventrales dont 4 restent face ventrale mais deux autres (Vs et V6) ont emigre face externe de la main. Les memes trichobothries se retrouvent chez Acanthothraustes brasiliensis (Fig. 6 et 10) ou 4 sont effectivement ventrales, les deux autres etant placees sur la face externe de la main. Les deux caracteres trichobothriotaxiques dont nous venons de parler sont si particu- lars au genre Euscorpius que nous devons admettre, sans reticence, la synonymie: Acan- thothraustes Mello-Leitao = Euscorpius Thorell. De plus, la possession de 4 trichobothries sur la face ventrale de la main (Fig. 6, 7 et 10), etant un caractere separant tres nettement Euscorpius flavicaudis de tous les autres Euscorpius, nous devons admettre la synonymie des deux espece s: Acanthothraustes brasiliensis (Mello-Leitao 1931), = Euscorpius flavicaudis (Geer 1778). Ajoutons, enfin, que le nombre et la position des trichobothries externes, du doigt fixe (Fig. 2, 3, 4, 5, 8 et 9), les trichobothries de la face externe de la main (Fig. 10 et 1 1), celles des faces interne et dorsale de l’avant-bras (tibia) (Fig. 12 et 13) et meme celles de la face ventrale (Fig. 16 et 17), sont identiques. Cette remarque est done un argument de plus pour admettre la synonymie proposee ci-dessus. L’etude comparee des autres caracteres conduit a la meme conclusion. Autres caracteres.— La morphologie du sternum et celle de l’opercule genital sont identiques chez les deux especes. Pour les peignes, Mello-Leitao (1931. 1945) precise 10-10 dents, e’est le nombre courant chez E. flavicaudis. 226 THE JOURNAL OF ARACHNOLOGY Les sillons et les carenes presentes sur le prosoma et les pedipalpes, correspondent entierement chez A. brasiliensis et chezE. flavicaudis ; la carene latero-dorsale existant sur la main se prolonge sur le doigt fixe ce qui est tres particulier au genre Euscorpius : cette disposition se retrouve chez les deux especes de meme que la presence, face interne du tibia des pedipalpes d’un tubercule basal tres developpe et spiniforme. Figs. 2-ll.-Pince gauche d ' Acanthothraustes brasiliensis et pince droite d’ Euscorpius flavicaudis: 2, A. brasiliensis, vue dorsale externe; 3, A. brasiliensis, vue dorsale; 4, E. flavicaudis, detail doigt; 5, E. flavicaudis, vue dorsale; 6, A. brasiliensis, vue ventrale; 1 ,E. flavicaudis, vue ventrale; 8, A. brasiliensis, vue interne; 9, E. flavicaudis, vue interne; 10, A. brasiliensis, vue externe; 11, E. flavicaudis, vue externe. Les abreviations designent les trichobothries. LOURENCO AND VACHON-ACANTHOTHRAUSTES BRASILIENSIS SYNONYMIE 227 Certes la coloration est tres estompee chez le type de A. brasiliensis, mais il est encore possible de constater sa ressemblance certaine avec celle de E. flavicaudis. Certains segments de la queue n’existent plus chez le type de A. brasiliensis , le telson par exemple; neanmoins, Mello-Leitao (1931) a donne une tres bonne photo du type et dans sa monographic (1945), il ajoute a nouveau une photo et en plus des schemas des peignes et des pedipalpes. Touts ces donnees permettent de lever nos doutes, si cela etait encore possible, quant a la similitude de A. brasiliensis et E. flavicaudis. Finalement la taille des specimens, bien que ce caractere soit dangereux a utiliser, est tres proche (45 mm donnes par Mello-Leitao en 1931) chez le type male de A. brasiliensis et chez les males de E. flavicaudis. REMARQUES ZOOGEOGRAPHIQUES Euscorpius flavicaudis est une espece habitant la region mediterraneenne dans le sud de la France (Vachon 1969). Toute possibility de sa presence dans la region amazonienne. Figs. 12-21 -Acanthothraustes brasiliensis et Euscorpius flavicaudis'. 12, A. brasiliensis, tibia gauche, vue dorsale; 13, E. flavicaudis, tibia droit, vue dorsale; 14, A. brasiliensis, tibia gauche, vue externe; 15, E. flavicaudis, tibia droit, vue externe; 16, A. brasiliensis, tibia gauche, vue ventrale; 17, E. flavicaudis, tibia droit, vue ventrale; 18, E. flavicaudis, femur droit, vue dorsale; 19, E. flavicaudis, femur droit, vue interne (detail); 20, E. flavicaudis, tarse de la patte ambulatoire; 21, E. flavicaudis, tranchant du doigt mobile droit. 228 THE JOURNAL OF ARACHNOLOGY Etat du Para, Bresil ne saurait etre retenue. Mello-Leitao (1931) mentionne “habitat: Para”; dans sa monographic (1945), il cite “habitat; Para, Belem.” L’addition du nom de la ville de Belem comme precision n’est pas justifiee, car sur l’etiquette accompagnant le type, se trouve uniquement l’indication “Para.” L’Etat du Para est une region tres vaste du Bresil, dont l’etendue depasse 1,2 millions de Km2 . L’un de nous (WRL) a eu la possibility d’effectuer des recherches dans cette region a plusieurs reprises; jamais aucun exemplaire, correspondant a A. brasiliensis n’a ete trouve. En outre, cette region, malgre sa grande surface a ete plus ou moins bien prospectee par d’autres chercheurs; aucun d’ entre eux n’a recolte un deuxieme exem- plaire de A brasiliensis’, le seul exemplaire connu est le type. II est important de remarquer que, dans l’Etat du Para on trouve des Chactidae, mais le genre Teuthraustes est present beaucoup plus a l’ouest, depuis le Venezuela jusqu’au Perou. Mello-Leitao n’etait pas un homme de terrain; il a toujours decrit ses especes a partir de specimens recoltes par quelqu’un d’autre que lui ou a partir d’un materiel existant dans les collections du Musee. Dans son travail (1931) il dit avoir trouve l’exemplaire de Teuthraustes brasilianus dans les collections du Museu Nacional; on peut admettre que cet exemplaire offert par un Musee europeen, mais non determine, aurait ete melange avec du materiel provenant de l’Etat du Para. D’ailleurs, dans ce meme travail (1931), Mello- Leitao decrit aussi une nouvelle espece, Tityus sampaiocrulsi, egalement de l’Etat du Para, avec la precision du Rio Cumina comme habitat. Nous croyons done que la proposition d’une nouvelle espece de Teuthraustes faite par Mello-Leitao provenait de sa faible connaissance du groupe. Par la suite, il se serait rendu compte que cette espece ( Teuthraustes brasilianus ) presentait des fortes differences par rapport aux autres especes du genre, mais n’ayant jamais etudie des Chactidae europeens, il a ete conduit a creer son nouveau genr e Acanthothraustes (Mello-Leitao 1945). REMERCIEMENTS Nous remercions ici bien vivement Mme le Dr. Anna Timotheo da Costa du Museu Nacional de Rio de Janeiro pour le pret du type de Acanthothraustes brasiliensis et Maurice Gaillard pour la realisation des dessins. TRAVAUX CITES Mello-Leitao, C. 1931. Dois novos escorpioes do Brasil. Bol. Mus. nac., 7(4):283-288. Mello-Leitao, C. 1932. Notas sobre escorpioes sul-americanos. Arq. Mus. nac., 34:1-46. Mello-Leitao, C. 1945. Escorpioes sul-americanos. Arq. Mus. nac., 40:1-468. Vachon, M. 1969. Nouvelles remarques sur la repartition en France metropolitaine du Scorpion mediterraneen Euscorpius flavicaudis (Geer) (Famille des Chactidae). Bull. Sci. Bourg., 26:189-202. Vachon, M. 1973. Etude des caracteres utilises pour classer les families et les genres de Scorpions (Arachnides). 1. La Trichobothriotaxie en Arachnologie. Sigles trichobothriaux et types de tricho- bothriotaxie chez les Scorpions. Bull. Mus. Nat. Hist. Nat., Paris, 3e ser. 140 (Zool. 104):857-958. Vachon, M. 1975. Recherches sur les Scorpions appartenant ou deposes au Museum d’Histoirenaturelle de Geneve. I. Contribution a une meilleure connaissance des especes et des sous-especes de Scor- pions du genre Euscorpius Thorell, 1876 (Fam. des. Chactidae). Rev. suisse Zool., 82(3):629-645. Manuscript received June 1980, accepted September 1980. Eberhard, W. G. 1981. The single line web of Phoroncidia studo Levi (Araneae, Theridiidae): A Prey attractant? J. Arachnol., 9:229-232. THE SINGLE LINE WEB OF PHORONCIDIA STUDO LEVI (ARANEAE: THERIDIIDAE): A PREY ATTRACTANT? William G. Eberhard1 Depto. de Biologia, Universidad de Valle, Cali, Colombia and Smithsonian Tropical Research Institute ABSTRACT Phoroncidia studo constructs snares consisting of single, more or less horizontal sticky lines. Fragmentary evidence suggests that the sciarid fly Bradysia sp. nr. coprophila may be attracted to these snares. INTRODUCTION Evolution toward reduced webs has occurred in several families of web building spi- ders. Among orb weavers, it appears to have been associated with the evolution of substances attractive to prey in Mastophora spp. (Eberhard 1977) (Araneidae), and prob- ably its near relatives, but not in Miagrammopes spp. (Lubin et al. 1978) (Uloboridae). In both of these groups as well as in Eucta sp. (Crome 1954) and Wixia sp. (Stowe 1978) (Araneidae), reduced webs are associated with particularly active attack behavior in which the spider either manipulates the web or snares the prey itself. Very simple webs which are presumably secondarily reduced have also been found in the theridiids Episinus spp. (Holm 1939, pers. obs.) and Phoroncidia spp. (= Ulesanis ) (Marples 1955). Episinus webs seem to be designed for pedestrian prey; they are often built at forks in branches and are somewhat similar to the “asterisk” webs of the araneid Wixia (Stowe 1978) except that they have adhesive at the ends of the lines. Phoroncidia webs differ in that they are strung between distant supports, with sticky portions away from the substrate rather than close to it (Marples 1955). Marples saw P. rotunda and P. quadratum webs consisting of only single threads partially covered with sticky material, and P. pukeiwa webs with from one to three lines with sticky segments. Sometimes P. pukeiwa manipulated its web by letting the line go slack when a prey was caught, but other times it did not. Marples did not give prey identities for any of these species. This note concerns the web of still another species in this genus, P. studo Levi. Fragmentary evidence is presented which suggests that this species, which manipulates its web only minimally during prey capture, attracts prey to its web. Resent address: S.T.R.I., and Escuela de Biologia, Universidad de Costa Rica, Cuidad Universitaria, Costa Rica. 230 THE JOURNAL OF ARACHNOLOGY STUDY SITE Observations were made of one adult female and probably four different immatures on 10-11 December 1976 in secondary forest at the edge of older forest near Yotoco, Valle, Colombia (el. about 1300 m, “very wet subtropical forest” Holdridgian life zone— Espinal and Montenegro 1963). The spiders were found in an area of relatively thick vegetation; the species was not common, and I could not find others on subsequent visits to the site. A voucher specimen is deposited in the Museum of Comparative Zoology, Cambridge, Mass. 02138. RESULTS The Web.— The three webs I saw were in place during the day and consisted of a single, more or less horizontal sticky line. One line (of an immature) was about 30 cm long, but only the 10 cm segment closest to the spider was sticky, while another (of a mature female) was sticky along almost its complete length of about 50 cm. Each spider rested near one end of its line, facing the sticky portion. I saw five spiders at night, but none had webs. It thus appears that the webs are normally up only during the day. Prey Capture.— I observed the behavior of two flies as they flew up to the web of a mature female and became trapped there. In both cases the fly hovered close to the web and made several darting flights toward it before finally touching it with its legs. In both cases I had a strong impression that the fly was not simply snared by the line as it flew by, but that it actively searched for it. Another indication of this was that all eight prey found in webs were held by the ends of their legs rather than by their wings or bodies. Once a prey hit the line, it struggled very little; it appeared that the glue held the flies so tightly that they could not move their legs. An adult female spider gave variable reactions to the arrival of prey. On three occa- sions she made no observable response and continued to wait until more prey arrived; once she immediately slackened the line and moved out toward the prey; and once she moved toward the prey without first slackening the line, and only gradually let it go slack as she approached the prey. After four flies had become trapped, this female left her resting site and attacked them. The spider evidently formed a bridge between the line in front and the line behind as described by Marples (1955), and as she moved she rolled up the sticky line and let out dry line behind. As she approached the first prey, she appeared to touch it several times with her front legs then turned 180° and wrapped it with alternating strokes of her hind legs. The prey was wrapped as it hung in the web, and was not spun or manipulated by the spider. After administering a brief wrapping (less than one minute), the spider appeared to give the prey a brief bite, and then moved on to the next, leaving the first suspended on the dry line she laid as she went. After subduing three of the four prey in the web in this manner, the spider moved back to the end under the leaf where she rested, letting out sticky silk behind as she went and thus replacing the segment of sticky line she had removed while attacking. As she moved she gathered up the immobilized prey and tightened the line. When she reached the end she attached the prey bundle to the line (?), turned 180° and apparently began to feed. She further tightened the web, just after turning, by slowly reeling in the line behind her with alternate pulls of her hind legs. The Prey.— At first I thought that the web probably functioned as a trap for the numerous small insects that were seen hanging on single silk lines in the vicinity (see EBERHARD-WEB OF PH OR ON CIDIA STUDO'. AN ATTRACTANT? 231 Lahmann and Zuniga in press , Eberhard 1980, Eberhard in press). A comparison between the spiders’ prey and the insects collected on such lines did not support this idea, how- ever. Of a total of 70 insects collected in the immediate area during the day, nearly all were Diptera (with two Hymenoptera and one Lepidoptera), and the large majority belonged to the gall gnat family Cecidomyidae. There was a relatively high diversity within this family, with males and females of many species present. On the other hand, all four prey specimens I collected were males of the sciarid fly Brady sia sp. nr. coprophila (Lintner) a species not present in the other collection (the other four prey I observed were also nematocerous flies, but were not collected; unfortunately I did not observe them with enough care to see if they were different.). Although the numbers are small, it appears that the prey did not represent a random sample of the insects resting on spider threads in that area. DISCUSSION Phoroncidia studo represents an extreme in web spiders with regard to its reduced attack behavior and reliance on its web for retention of prey that have been intercepted. It contrasts sharply in this respect with other theridiids and araneids with reduced webs. It is also apparently unique in replacing its sticky line immediately after attacking prey. It seems to be similar to the Phoroncidia species studied by Marples (1955) with respect to web form, attack behavior, and web tightening behavior. The behavior of prey arriving at webs, the fact that all prey were caught by the ends of their long legs, and the fact that only males of a single species were captured all suggest that P. studo uses some sort of attractant to lure flies to its web. M. Robinson has also pointed out to me that the fact that the spiders allow several prey to accumulate before attacking is logical in terms of the attractant hypothesis: if the web is particularly costly (contains attractant) and the likelihood of prey arriving is good (they are attracted), then infrequent web destruction and replacement would be advantageous. Dinopus longipes which hunts by ant trails (prey arrive often) seems to behave just as P. studo , allowing more than one small ant to accumulate in its web before collapsing the web and wrapping the prey (Robinson and Robinson 1971). It is not known whether male Bradysia sp. nr. coprophila respond to chemical stimuli. Obviously more work on this or another, commoner species would be of interest. ACKNOWLEDGMENTS Dr. H. W. Levi kindly identified the spider, and Drs. R. Gagne and W. A. Steffans the fly. The study was conducted during a field trip led by Drs. Jorge E. Orejuela and Humberto Alvarez, and was supported by the Comite de Investigaciones, Universidad del Valle, Cali, Colombia. REFERENCES CITED Crome, W. 1954. Breschriebung, Morphologie und Lebensweise der Eucta kaestneri sp. n. (Araneae, Tetragnathidae). Zool. Jahrb., 82(5):425-452. Eberhard, W. G. 1977. Aggressive chemical mimicry by a bolas spider. Science, 198:1173-1175. Eberhard, W. G. 1980. Spider and fly play cat and mouse. Nat. Hist., 89( 1 ) :56-6 1 . Eberhard, W. G. in press. Argyrodes attenuatus (Theridiidae): a web that is not a snare. Psyche. 232 THE JOURNAL OF ARACHNOLOGY Espinal, L. S. and E. Montenegro. 1963. Formaciones Vegetales de Colombia. Inst. Geogr. Agustm Codazzi, Bogota, 201 p. Lahmann, E. J. and C. M. Zuniga. 1981. Use of spider threads as resting places by tropical in- sects. J. Arachnol., in press. Lubin, Y. D., W. G. Eberhard, and G. G. Montgomery. 1978. Webs of Miagrammopes (Araneae: Ulobridae) in the neotropics. Psyche, 85(1 ) : 1-23. Holm, A. 1939. Beitrage zur Biologie der Theridiiden. Festr. Strand., 5:56-67. Marples, B. J. 1955. A new type of web spun by spiders of the genus Ulesanis with the description of two new species. Proc. Zool. Soc. London, 125:751-760. Robinson, M. and B. Robinson. 1971. The predatory behavior of the ogre-faced spider Dinopis longipes F. Cambridge. Am. Midi. Nat., 85:85-96. Stowe, M. 1978. Observations of two nocturnal orbweavers that build specialized webs: Scoloderus cordatus and Wixia ectypa (Araneae: Araneidae). J. Arachnol., 6:141-146. Manuscript received May 1980, revised July 1980. THE AMERICAN ARACHNOLOGICAL SOCIETY President: Herbert W. Levi (1979-1981) Museum of Comparative Zoology Harvard University Cambridge, Massachusetts 02138 Membership Secretary: Norman I. Platnick (appointed) American Museum of Natural History Central Park West at 79th Street New York, New York 10024 Secretary: William A. Shear (1980-1982) Department of Biology Hampden-Sydney College Hampden-Sydney, Virginia 23943 President-Elect: Jonathan Reiskind (1979-1981) Department of Zoology University of Florida Gainesville, Florida 32601 Treasurer : Norman V. Horner (1979-1981) Department of Biology Midwestern State University Wichita Falls, Texas 76308 Directors: Charles D. Dondale (1979-1981) Susan E. Riechert (1979-1981) Vincent D. Roth (1980-1982) The American Arachnological Society was founded in August, 1972, to promote the study of the Arachnida, to achieve closer cooperation between amateur and professional arachnologists, and to publish The Journal of Arachnology. Membership in the Society is open to all persons interested in the Arachnida. Annual dues are $20.00 for regular members, $12.00 for student members. Correspondence con- cerning membership in the Society must be addressed to the Membership Secretary. Members of the Society receive a subscription to The Journal of Arachnology. In addi- tion, members receive the bi-annual newsletter of the Society, American Arachnology. American Arachnology , edited by the Secretary, contains arachnological news and comments, requests for specimens and hard-to-find literature, information about arach- nology courses and professional meetings, abstracts of the papers presented at the Society’s meetings, address changes and new listings of subscribers, and many other items intended to keep arachnologists informed about recent events and developments in arach- nology. Contributions for American Arachnology must be sent directly to the Secretary of the Society. The Eastern and Western sections of the Society hold regional meetings annually, and every three years the sections meet jointly at an International meeting. Information about meetings is published in American Arachnology , and details on attending the meetings are mailed by the host(s) of each particular meeting upon request from interested persons. The next International meeting will be held during 5-7 August 1981, hosted by Dr. Susan E. Riechert, Department of Zoology, University of Tennessee, Knoxville, TN 37916, USA. CONTENTS THE JOURNAL OF ARACHNOLOGY VOLUME 9 SPRING 1981 NUMBER 2 Feature Articles The ambypygid genus Phrynus in the Americas (Amblypygi, Phrynidae), Diomedes Quintero 117 The erigonine spiders of North America. Part 3. The genus Scotinotylus Simon (Araneae, Linyphiidae), A F. Millidge 167 Water and metabolic relations of cave-adapted and epigean lycosid spiders in Hawaii, Neil F. Hadley, Gregory A. Ahearn and Francis G. Howarth 215 Complements a la description d ' Acanthothraustes brasiliensis (Mello-Leitao 1931) (= Teuthraustes brasilianus , Mello-Leitao 1931), synonyme d 'Euscorpius flavicaudis (Geer 1778) (Scorpiones, Chactidae), Wilson R. Lourengo and Max Vachon 223 The single line web of Phoroncidia studo Levi (Araneae, Theridiidae): A prey attractant? William G. Eberhard 229 Cover illustration, Schizomus pentapeltis (Cook), by W. David Sissom Printed by The Texas Tech University Press, Lubbock, Texas Posted on July 1981 The Journal of ARACHNOLOGY OFFICIAL ORGAN OF THE AMERICAN ARACHNOLOGICAL SOCIETY VOLUME 9 FALL 1981 NUMBER 3 THE JOURNAL OF ARACHNOLOGY EDITOR’. Oscar F. Francke, Texas Tech University. ASSOCIATE EDITOR ’. B. J. Kaston, San Diego State University. EDITORIAL ASSISTANT: James C. Cokendolpher, Texas Tech University. EDITORIAL BOARD: Charles D. Dondale, Agriculture Canada. William G. Eberhard, Universidad de Costa Rica. Maria E. Galiano, Museo Argentino de Ciencias Naturales. Willis J. Gertsch, American Museum of Natural History. Neil F. Hadley, Arizona State University. B. J. Kaston, San Diego State University. Herbert W. Levi, Harvard University. Emilio A. Maury, Museo Argentino de Ciencias Naturales. William B. Muchmore, University of Rochester. Martin H. Muma, Western New Mexico University. William B. Peck, Central Missouri State University. Norman I. Platnick, American Museum of Natural History. Susan E. Riechert, University of Tennessee. Michael E. Robinson, Smithsonian Tropical Research Inst. Jerome S. Rovner, Ohio University. William A. Shear, Hampden-Sydney College. Carlos E. Valerio, Universidad de Costa Rica. Stanley C. Williams, San Francisco State University. THE JOURNAL OF ARACHNOLOGY is published in Winter, Spring, and Fall by The American Arachno logical Society in cooperation with The Graduate School, Texas Tech University. Individual subscriptions, which include membership in the Society, are $20.00 for regular members, $12.00 for student members. Institutional subscriptions to The Journal are $25.00. Correspondence concerning subscription and membership should be addres- sed to the Membership Secretary (see back inside cover). Back issues of The Journal are available from Dr. William B. Peck, Department of Biology, Central Missouri State Univer- sity, Warrensburg, Missouri 64093, U.S.A., at $5.00 for each number. Remittances should be made payable to The American Arachnological Society. Change of address notices must be sent to both the Secretary and the Membership Secretary. Manuscripts for THE JOURNAL OF ARACHNOLOGY are acceptable only from cur- rent members of the Society, and there are no page charges. Manuscripts must be type- written double or triple spaced on 8.5 in. by 11 in. bond paper with ample margins, and may be written in the following languages: English, French, Portuguese, and Spanish. Contributions dealing exclusively with any of the orders of Arachnida, excluding Acari, will be considered for publication. Papers of a comparative nature dealing with che Iter- ates in general, and directly relevant to the Arachnida are also acceptable. Detailed instructions for the preparation of manuscripts appear in the Fall issue of each year, and can also be obtained from the Editor and the Associate Editor. Manuscripts that do not follow those instructions will be returned to the author(s) without the benefit of review. Manuscripts and all related correspondence must be sent to Dr. B. J. Kaston, 5484 Hewlett Drive, San Diego, California 92115, U.S.A. Francke, O. F. and M. E. Soleglad. 1981. The family Iuridae Thorell (Arachnida, Scorpiones). J. Arachnol., 9:233-258. THE FAMILY IURIDAE THORELL (ARACHNIDA, SCORPIONES) Oscar F. Francke Departments of Biological Sciences and Entomology, and The Museum, Texas Tech University, Lubbock, Texas 79409 and Michael E. Soleglad 1 502 Dupont Drive Lemon Grove, California 92045 ABSTRACT The phylogenetic relationships of some chactoid genera are analyzed. The number of lateral eyes, the only character used to separate the families Chactidae and Vaejovidae, is rejected because of considerable intrageneric, intergeneric, and interfamilial variability. The value of trichobothrial num- bers and patterns in scorpion systematics is discussed, and the terminologies proposed by both Stahnke and Vachon are considered to be of limited value in phylogeny reconstruction because they postulate incongruent homologies. The following sister-group relationships and monophyletic taxa are hypothesized: Caraboctonus + Hadruroides = Caraboctonini, new; Hadrurus = Hadrurini, new; Cara- boctonini + Hadrurini = Caraboctoninae; Calchas + Iurus = Iurinae; Caraboctoninae + Iurinae = Iuridae, new rank. A key to scorpion families, and a key to subordinate taxa of Iuridae are presented. The suprageneric taxa of iurids are diagnosed and compared. A vicariance model is presented to explain the present geographical distribution of the family. INTRODUCTION “Considering the wide geographical range of this group [Iurini]and the difference of aspect presented by such of its members as Scorpiops and Hadrurus, one would be inclined to think the assemblage an unnatural one. But the intermediate forms that exist seem to show that this is not the case. For instance, from Scorpiops to Iurus is not a great leap; and similarly we can proceed from Iurus through Uroctonus and Anuroctonus to Vejovis, or through Hadruroides and Caraboctonus to Hadrurus. Hadrurus undoubtedly differs very much from Iurus, but no one will probably dispute that it is nearly allied to Caraboctonus', and the similarity that obtains between Caraboctonus and Iurus with respect to armature of the mandible, the hairy clothing of the soles of the feet, &c., may surely, when taken in conjuction with the other features already pointed out as characteristic of the Iurini, point to real kinship between the two.” Pocock 1893:309 234 THE JOURNAL OF ARACHNOLOGY Pocock’s (1893) subfamily Iurini (sic), its characterization and its included genera, represent (with additional genera described since) what is known as the family Vaejovidae of recent authors (cf. Stahnke 1974). Considerable evidence, however, clearly indicates that “the assemblage is an unnatural one.” Some of that evidence and the taxonomic changes dictated by it are given below. The genera discussed by Pocock and whose relationships form the central theme of this contribution presently constitute three distinct subfamilies within the Vaejovidae: Caraboctonus Pocock and Hadruroides Pocock make up the Caraboctoninae Kraepelin, Hadrurus Thorell and Anuroctonus Pocock form the Hadrurinae Stahnke, and Iurus Thorell constitutes the monogeneric Iurinae Thorell. To complicate matters further, the chactid Calchas Birula must also be considered, for as Vachon (1971:718) indicates: “La trichobothriotaxie, la denture des cheliceres dif- ferent beaucoup de celles que Ton retrouve chez les autres Chactidae, ce qui prouve l’originalite de cette sous-famille (Calchinae Birula) a l’interieur des Chactidae. Par contre ces caracteres sont identiques a ceux possedes par le genre Iurus Thorell, appartenant a la famille des Vejovidae . . .” Thus, the relationships of four subfamilies (in two different families) are examined in this contribution. Cladistic arguments are presented whenever possible. Seven Recent families of scor- pions are currently recognized. The family Buthidae is apparently monophyletic based on trichobothrial patterns (Vachon 1974), cheliceral dentition (Vachon 1963), and male and female reproductive systems (Francke 1979), and may be the sister group of all the other Recent scorpion families (see also Lamoral 1980). The monogeneric Chaerilidae is mono- phyletic based on the trichobothrial pattern (Vachon 1974), chelicerae (Vachon 1963), and gnathobase morphology. Based on the male and female reproductive systems it appears that the Chaerilidae represent the sister-group of the remaining five Recent families. The superfamily Scorpionoidea (Diplocentridae plus Scorpionidae) appears monophyletic (with the exclusion of one genus perhaps) based on the female reproduc- tive system and method of embryonic nutrition. The family Bothriuridae, classified by some in its own superfamily and by others in the Chactoidea, also appears monophyletic based on the female reproductive system, sternum shape, and venom gland morphology. Finally, the families Chactidae and Vaejovidae are here not considered to be monophy- letic, although as a higher taxon (superfamily Chactoidea) they might be. The families Chactidae and Vaejovidae are separated only by the number of lateral eyes: Chactidae with two pairs and Vaejovidae with three. This character, however, is known to be unreliable. Among “chactids” Broteochactas Pocock and Teuthraustes Simon actually have three or four pairs, Broteas Koch has two, three or four pairs, and Chactopsis Kraepelin has three pairs (Gonzales 1977); and the Typhlochactini Mitchell has five eyeless species in two genera (Francke 1981). Among “vaejovids” there is also considerable variability (Gertsch and Soleglad 1972); Anuroctonus has four pairs, and Parascorpiops Banks has two pairs (Francke 1976). This variability has resulted in con- siderable taxonomic confusion over the years: for example, the genus Uroctonoides Chamberlin was described as a vaejovid and subsequently shown to be a junior synonym of the chactid Teuthraustes (Soleglad 1973); and the genus Chaerilomma Roewer was described as a chactid and later shown to be a junior synonym of the vaejovid Iurus (Vachon 1966). The chaos in classification resulting from the use of this unreliable character is such that taxonomists must often identify the genus first, and then find out to which family it is assigned. FRANCKE AND SOLEGLAD-THE FAMILY IURIDAE THORELL 235 The phylogenetic relationships of chactoid scorpions can only be understood after monophyletic taxa are recognized, and the sister-group relationships among them figured out. Consequently, outgroup comparisons for this study were extremely difficult, as the level of generality of a given character was nearly impossible to determine. Comparisons are made within (a) the chactoids, (b) chactoids and bothriurids, (c) chactoids, bo- thriurids, and scorpionoids, (d) all non-buthid Recent scorpions, and finally (e) among all Recent scorpions. The postulated polarity of the transformation series can thus be correct at one level of generalization, and incorrect at another; only future studies will tell. The aim of this contribution (and others in preparation) is to hypothesize monophy- letic groupings within the chactoids: (a) by showing that previous hypotheses are not supported by the data available, and (b) by identifying shared derived characters (= synapomorphies) defining some chactoid elements. PHYLOGENETIC RELATIONSHIPS The relationships of Caraboctonus and Hadruroides.— The genus Caraboctonus con- tains a single species, C. keyserlingi Pocock, from central Chile, and Hadruroides contains seven species from Peru and Ecuador, including the Galapagos Islands (Maury 1975, Francke and Soleglad 1980). These two genera share a number of unique features which corroborate their hypothesized sister-group relationship (by being the only two genera placed in the Caraboctoninae). The most significant synapomorphies, based on outgroup comparisons with all Recent scorpions, are the trichobothrial patterns (Vachon 1974, Maury 1975), and the lack of a well developed ventral median claw (= unguicular spine) on the tarsus. The dentition pattern on the pedipalp chela fingers is basically the same on both genera, with six (fixed finger) or seven (movable finger) distinct oblique rows. Had- ruroides differs from Caraboctonus , however, in the presence of internal and external supernumerary granules flanking the oblique rows of principal granules. The supernu- merary granules appear gradually after the third or fourth molts, and are absent on all other Recent scorpions with similar dentition patterns (i.e., six or seven distinct oblique rows). Thus, by outgroup comparison and by ontogenetic arguments (Nelson 1978), supernumerary granules are autapomorphic for Hadruroides. The sternum can be about as long as wide with a deep longitudinal furrow in Had- ruroides, or much wider than long and at most with a deep pit posteriorly and no furrow in Caraboctonus. Among chactoids the sternum is usually as long as wide with a longi- tudinal furrow. Therefore, the character state in Caraboctonus is hypothesized, by out- group comparison, to represent an autapomorphy. The relationships of Anuroctonus and Hadrurus— Stahnke (1974) included only Had- rurus, with eight species, and Anuroctonus, with one species, in the subfamily Hadrurinae Stahnke. “This subfamily [Hadrurinae] is primarily characterized by its large number of trichobothria (86 to 145) as compared with all the other subfamilies [of Vaejovidae] (45 to 63) . . . Although the two genera assigned to this subfamily vary widely in some respects, the trichobothria indicate a closer affinity between them than to the genera herein assigned to the Vaejovinae ” (Stahnke 1974:116, 11 8). The implied hypothesis is clear: these two genera are more closely related to each other than either is to any other Recent scorpion. The validity of this hypothesis is examined next. 236 THE JOURNAL OF ARACHNOLOGY TRICHOBOTHRIAL NUMBERS. Stahnke’s use of absolute trichobothrial numbers on the pedipalp chela and tibia to characterize the Hadrurinae is somewhat superficial and vague. First, if Stahnke (1974, table 1, p. 121) used trichobothrial numbers as a measure of phenetic distance, we observe that he characterizes Hadrurus as having 145 trichobo- thria and Anuroctonus as having 86, a difference of 59. Scorpiops Peters varies from 53 to 64 trichobothria, a minimal difference of only 22 with respect to Anuroctonus. Phe- netically at least, it would seem that Anuroctonus is closer to Scorpiops than either is to Hadrurus. Second, we might consider that Stahnke implied a cladistic interpretation, i.e., that the increase in trichobothrial count (to 86 or higher) is a synapomorphy. The genus Dasyscorpiops Vachon (Vaejovidae, Scorpiopsinae), recently described from Malacca, has 1 12 trichobothria (Vachon 1974). Cladistically, however, Dasyscorpiops is closer to Scor- piops than it is to either Anuroctonus or Hadrurus (based on cheliceral dentition, pedi- palp finger dentition, carapace shape, and trichobothrial pattern). Therefore, a mere increase in trichobothrial numbers (to 86 or higher) does not necessarily represent a synapomorphy among vaejovids. Next we might examine which regions of the pedipalps bear “larger than average numbers” of trichobothria in Hadrurus and Anuroctonus with respect to other vaejovids. These regions are the ventral aspect of the chela, and the external and ventral aspects of the tibia. Among vaejovids, all members of the subfamily Scorpiopsinae have “supernu- merary” trichobothria on the external and ventral aspects of the tibia, and some Scorpiops also have them on the chela (Vachon 1974); Uroctonus bogerti Gertsch and Soleglad (Vaejovinae) and Paravaefovis pumilis (Williams) (Vaejovinae) have them on the ventral aspect of the chela. Thus, Hadrurus and Anuroctonus are not unique among vaejovids in this respect. As a matter of fact, according to Vachon (1974, table 4, pp. 935-936), “supernumerary” trichobothria occur in over 50% of the non-buthid genera studied by him. The presence of “supernumeraries” on both the tibia and chela occur in three genera of Bothriuridae, one genus of Chactidae, and seven genera of Scorpionidae, in addition to the vaejovids under consideration. Therefore, regardless of whether we examine Stahnke’s use of trichobothrial numbers phenetically or cladistically, and in the latter case whether we examine total or region- alized counts, the data do not offer strong support of the hypothesized sister-group relationship between Anuroctonus and Hadrurus. TRICHOBOTHRIAL PATTERNS. Vachon (1972, 1974) has indicated that close examination of the positions of individual trichobothria is a more fruitful approach to the study of phylogenetic relationship in scorpions than merely counting them. Particularly interesting and relevant for this contribution are his illustrations of the patterns in Iurus and Calchas, reproduced in Figs. 1-8. Two systems of trichobothrial nomenclature in scorpions are in use, Vachon’s (1974) and Stahnke’s (used in various publications; for the present discussion reference is make to his 1974 contribution on vaejovids). Unfortunately both systems sometimes generalize unnecessarily in their attempt to find “universal” patterns, and therefore sometimes use the same designation for non-homologous trichobothria or different designations for homologous ones. Vachon (1974) recognized three “universal” patterns among scorpions. Pattern “A” for the Buthidae, “B” for Chaerilidae, and “C” for all other Recent families. These patterns are characterized by a given number of trichobothria on the femur, tibia, and chela of the pedipalps; the “universal” or “fundamental” number he termed orthobo- thriotaxia, and deviations from this number are designated neobothriotaxia (the evolu- FRANCKE AND SOLEGLAD-THE FAMILY IURIDAE THORELL 237 tionary implications of this terminology are of no concern here). Deviations resulting in a greater number of trichobothria are termed “additive neobothriotaxia” {neobothriotaxie majorante), and decreases are “substractive neobothriotaxia” (< neobothriotaxie mino- rante ). Of 55 genera with pattern “C” examined by Vachon, only 26 (47%) are orthobo- thriotaxic, and 29 (53%) have additive neobothriotaxia. Thus, the “universality” of orthobothriotaxia “C” should be questioned on this basis alone. Cladistic hypotheses that orthobothriotaxia “C”, established on the basis of absolute numbers, is either apomor- phic or plesiomorphic are rejected because the patterns (i.e., the relative positions of the trichobothria) differ considerably between groups. It is important to note, however, that trichobothria can be gained or lost, and the direction of evolutionary change (i.e., the 5 6 7 8 Figs. l-8.-Trichobothrial patterns of Calchas Birula (Chactidae) and Iurus Thorell (Vaejovidae): 1-4, Calchas ; 5-8, Iurus. 1, 5, external aspect of pedipalp chela; 2, 6, ventrointernal aspect of chela; 3, 7, external aspect of pedipalp tibia; 4, 8, ventral aspect of tibia (from Vachon 1974). 238 THE JOURNAL OF ARACHNOLOGY polarity of the transformation series) must be empirically established for each taxon (see Francke 1981). The trichobothrial patterns of the chela of Hadruroides lunatus (Koch), Hadrurus pinteri Stahnke, Anuroctonus phaiodactylus (Wood), and Iurus dufoureius (Brulle), with some of the homologies postulated by Vachon (1974) and Stahnke (1974) are shown in Figs. 9-16. In Vachon’s scheme (Figs. 9-12), trichobothria Db-Dt are on the external aspect of the fixed finger in Hadruroides and Caraboctonus, basally on the dorsoextemal aspect of the palm in Hadrurus and Anuroctonus , and Dt is distally on the external aspect of the palm in Iurus. Since other vaejovid genera, such as Uroctonus Thorell, Scorpiops Peters, and Vaejovis Koch, have trichobothria Db-Dt basally on the palm, Vachon (1974:898) explains the condition observed in Hadruroides and Caraboctonus as being the result of “phylogenetic displacement” (or “emigration” in short form): “Nous pou- vons logiquement admettre ces deplacements puisque lesdites trichobothries son absentes a la base de la main et qu’il en existe 2 de plus sur le doigt fixe.” Stahnke’s scheme (Figs. 13-16) also implies considerable trichobothrial movement, the origin of which is never addressed. Trichobothrium/2 is dorsally on the base of the fixed finger on Caraboctonus and Hadruroides and Hadrurus , is nonexistant on Anuroctonus , and occurs distally on the internal aspect of the finger in Iurus. Comparing the patterns of Iurus and Calchas (Figs. 1-8), one observes a more credible form of trichobothrial displacement; namely, that the trichobothria on the Finger are rather equidistant on both genera, but on Calchas they cover most of the finger while on Iurus they cover the distal one-half to two-thirds of the finger only. In this instance, allometric growth of the base of the finger of Iurus could account for differences observ- ed. The mechanism(s) whereby trichobothria can “emigrate” is left unexplained by both Vachon and Stahnke. Trichobothria are mechanoreceptors innervated by a single bipolar neuron each, and any mechanism proposed to account for trichobothrial migration must also adquately explain the migration of their respective neurons. Indiscriminate migration of trichobothria is one possible explanation, albeit a rather unconvincing one, for the existence of certain preconceived patterns. An alternative explanation is that there are no “universal” patterns and that trichobothrial migration is minimal at best. We hypothesize that trichobothria occupying similar positions are homologous, and that discrepancies in Vachon’s and Stahnke’s topologies (determined in large part by their terminologies) are due to gain or loss of trichobothria either by suppression of trichobothrial development, by the development of a different type of sensory seta in that position, or by some other means (see Francke 1981). There is no evidence of trichobothrial migration, and the alleged displacements may merely reflect the shortcomings of the terminologies developed by Vachon and Stahnke, respectively. Evidence of trichobothrial gain or loss is widespread. First, at the individual level it is not uncommon to find asymmetrical specimens with an “aberrant” number of trichobothria on one pedipalp (Maury 1973, Vachon 1975). These “aberrations” involve the loss of one or more trichobothria in some cases, the gain of trichobothria in others, and in many instances it is difficult to determine the direction of change. Second, at the species level considerable variability exists among taxa with “additive neobothriotaxia.” Excellent examples are found in the chactids Euscorpius spp. (Curcic 1972, Vachon 1975, Valle 1975) and Megacormus spp. (Soleglad 1976a), bothriurids Brachistostemus spp. (Maury 1973), scorpionids Urodacus spp. (Koch 1977), and vaejovids Hadrurus spp. (Soleglad 1976b). Third, at the generic level considerable variability is also found FRANCKE AND SOLEGLAD-THE FAMILY IURIDAE THORELL 239 among closely related species. The five genera cited above, considered monophyletic by most workers, are excellent examples. Soleglad’s (1976b) work on Hadrurus is particu- larly relevant to this contribution: some species have no “accessory” (= supernumerary ?) trichobothria on the chela, some have them on the internal aspect of the fixed finger base, still others have them on the external aspect of the palm, and some even have them on both areas. Noteworthy is Hadrurus pinteri Stahnke, which has an “accessory” tricho- bothrium on the external aspect of the pedipalp finger (its significance will be discussed later). Why, then, are authors reluctant to consider trichobothrial gain or loss as an alterna- tive explanation to “migration” in their search for homologies between genera? Thus, according to Vachon (Figs. 9-12) trichobothrium Dt , located sub medially on the dorsal aspect of the palm in Hadrurus , has migrated (with its “companion” Db ) to a submedian position on the external aspect of the finger in Hadruroides , and to a subdistal position on the external aspect of the palm in Iurus. We find this hypothesis entirely unparsimoni- ous: Hadrurus pinteri and Hadruroides plus Caraboctonus have six trichobothria on the external aspect of the finger, and these we consider homologous; the differences observed on the dorsal aspect of the palm are due to gain or loss of trichobothria in either taxon (once sister groups relationships are established for the taxa in question, cladistic methods can be used to formulate hypotheses about the direction of change). Likewise, as indi- cated above some Hadrurus species have “accessory” trichobothria on the external aspect of the palm, and the position of one of these corresponds very well with the position of Dt in Iurus , and is hence presumably homologous to Dt. The shortcomings of Vachon’s terminology are illustrated by the fact that trichobo- thrium Db in Hadruroides and Caraboctonus is in the same position as eb in Hadrurus, and dsb in Anuroctonus (Fig. 9-12). Stahnke’s termonology is not much better: position- ally, M2 in Hadruroides and Caraboctonus = Mx in Hadrurus = D6 in Anuroctonus, and Mi in Hadruroides and Caraboctonus = I2 in Hadrurus = Z)4 in Anuroctonus (Figs. 13-16). In conclusion, trichobothrial numbers and trichobothrial patterns need to be carefully analyzed before being used to postulate taxonomic “affinities” among scorpion taxa. Stahnke’s use of absolute trichobothrial numbers to characterize his subfamily Had- rurinae is unjustified. The fact that Hadrurus and Anuroctonus used to be included in the Vaejovinae before Stahnke’s action is no reason to restrict comparisons to those genera assigned to it. As early as 1893 Pocock indicated that “no one will probably dispute that it [Hadrurus] is nearly allied to Caraboctonus .” CHELICERAL DENTITION. Hadrurus, Hadruroides, and Caraboctonus posses a well developed tooth on the inferior border of the movable finger of the chelicerae (Figs. 17, 18). Among all other non-buthid Recent scorpions this tooth is only found in Iurus and Calchas (Figs. 19, 20). Anuroctonus sometimes has a small denticle on this border (Fig. 21), but we do not consider it homologous to the large tooth mentioned above. Denticles on the inferior border of the movable finger of the chelicerae occur also in the Scor- piopsinae, and in Paruroctonus (Fig. 22), Uroctonus, and Vaejovis among the Vaejovinae. The condition observed in Anuroctonus is here hypothesized to be homologous to that of vaejovines, and that of Hadrurus to be homologous to Hadruroides and Caraboctonus. VENOM GLANDS. Pavlovsky (1924a) compared the morphology of venom glands in several scorpion genera, and recognized two basic types: Type I (Primitive) glands are unfolded, sac-like merocrine organs, and Type II (Complex) glands have one or more folds involving both the secretory epithelium and the basal membrane. Among the taxa 240 THE JOURNAL OF ARACHNOLOGY we are concerned with, Pavlovsky reported Type I venom glands in Uroctonus, Scorpiops, and Vaejovis, and Type II venom glands in Caraboctonus, Hadruroides, Hadrurus, and Iurus. Studies in progress on numerous scorpion taxa have confirmed Pavlovsky’s findings and extended them. Among vaejovids Type I venom glands are present in Anuroctonus, Figs. 9- 12. -Diagrammatic illustrations of trichobothrial patterns of pedipalp chela showing homol- ogies postulated by Vachon (1974): 9, Hadruroides lunatus (Koch); 10, Hadrurus pinteri Stahnke; 11, Anuroctonus phaiodactylus (Wood); 12, Iurus dufoureius (Brulle). KEY: ♦ = dorsal series of fixed finger, * = external series of fixed finger, • = Dorsal series of palm, ■ = External terminal series of palm, o = other trichobothria whose particular designations are not relavant here. FRANCKE AND SOLEGLAD-THE FAMILY IURIDAE THORELL 241 Paruroctonus, Paravaejovis Williams, Serradigitus Stahnke, and Vejovoidus Stahnke. Embryological studies (Pavlovsky 1924a, Probst 1972) have shown that folded glands are indeed derived from simple unfolded glands. Thus, we hypothesize that Caraboctonus, Hadruroides, and Hadrurus share a derived character state not found in Anuroctonus. Figs. 13-16. -Diagrammatic illustrations of trichobothrial patterns of pedipalp chela showing homologies postulated by Stahnke (1974): 13, Hadruroides lunatus (Koch); 14, Hadrurus pinteri Stahnke; 15, Anuroctonus phaiodactylus (Wood); 16, Iurus dufoureius (Brulle). KEY: ♦ = Internal series of fixed finger, * = Dorsal series of fixed finger, ■ = Median series of chela, • = Basal series of palm, o = other trichobothria whose particular designations are not relevant here. 242 THE JOURNAL OF ARACHNOLQGY HEMISPERMATOPHORES. The spermatophore of Hadrurus arizonensis Ewing was recently described and noted to differ significantly from all described scorpion spermato- phores (Francke 1979). Given the difficulties of obtaining scorpion spermatophores, several authors have opted to dissect and examine the hemispermatophores before they are deposited by males (e.g., Vachon 1952, Maury 1976, Koch 1977). The hemispermatophores of Hadrurus (Figs. 23-26), Hadruroides (Figs. 27-33), and Caraboctonus (Figs. 34-37), are characterized by the absence of a capsule and the absence of a distinct truncal flexure. Their “triggering” mechanism, i.e., the action that enables Figs. 17-22. -Ventral aspect of chelicera, showing dentition on both fixed and movable fingers. Note the conspicuous, large tooth (darkened) on the ventral aspect of the movable finger on figs. 17-20: 17, Hadrurus arizonensis Ewing; 18, Caraboctonus keyserlingi Pocock; 19 , lurus dufoureius (Brulle); 20, Calchas nordmanni Birula; 21, Anuroctonus phaiodactylus (Wood); 22, Paruroctonus gracilior (Hoffmann). FRANCKE AND SOLEGLAD-THE FAMILY IURIDAE THORELL 243 females to retrieve sperm (going from the preinsemination to the postinsemination state of the spermatophore), apparently resides in the broad flange that marks the mid-dorsal aspect of the hemispermatophore. [The genital operculi of the female separate and engage the flange on each side of the spermatophore.] The rounded, “free” lobes medial to the flange are presumably involved in the formation of the sperm duct which directs sperm from the trunk of the spermatophore into the genital opening of the female. A possible mechanism for sperm ejection in the absence of a truncal flexure has been proposed elsewhere (Francke 1979). The hemisphermatophore of Anuroctonus (Figs. 38-41) has a well developed truncal flexure. A capsule, however, is apparently lacking, or if present consists of non-sclerotized membranes which could not be discerned by the techniques used in this study. Note- worthy are: the truncal flexure; the abrupt transition from trunk to lamella, marked by a pronounced inflection; and the small tooth mid-dorsally, which presumably engages the female genital operculum during sperm transfer. The hemispermatophores of Vaejovis (Figs. 42-45), Uroctonus (Figs. 46-49), and Paruroctonus (Figs. 50-52), possess well developed truncal flexures and heavily sclero- tized capsules. The abrupt transition from trunk to lamella with its pronounced inflec- tion, and the mid-dorsal tooth, are similar (and perhaps homologous) to those in Anuroctonus. Spermatophores and hemisphermatophores with well developed truncal flexures occur in bothriurids, scorpionids, diplocentrids, and some chactids in addition to the vaejovids cited above. A truncal flexure and a capsule are absent in buthid and chaerilid hemisper- matophores, and their absence outside these groups is therefore considered plesimorphic Figs. 23-26.— Hemispermatophore of Hadrurus arizonensis Ewing: 23, external aspect (dorsal re- gion to the right); 24, dorsal aspect; 25, internal aspect; 26, ventral aspect. (Total length 10.0 mm) 244 THE JOURNAL OF ARACHNOLOGY (more general). The absence of a distinct truncal flexure and a capsule in the hemisper- matophores of Caraboctonus, Hadruroides, and Hadrurus is therefore uninformative with respect to their phylogenetic relationships. SPERMATOZOAN AXONEME. Jespersen and Hartwick (1973) studied the fine struc- ture of spermiogenesis in some North American vaejovids. They found that the axonemal structure of sperm in Hadrurus is of a 9 + 1 pattern, whereas Vaejovis, Uroctonus, and Anuroctonus have a 9 + 0 pattern. They further indicate that a 9 + 2 pattern, as found in Euscorpius Thorell (Chactidae) is plesimorphic, and that the 9 + 1 and 9 + 0 patterns represent derived character states. Additional studies, especially of chactoids, are needed before phylogenetic hypotheses based on this character can be formulated. Nonetheless, this character fails to support Stahnke’s hypothesized sister-group relationship between Anuroctonus and Hadrurus. We have been unable to find any synapomorphies between Anuroctonus and Hadrurus that would corroborate the validity of Stahnke’s subfamily Hadrurinae, and have found synapomorphies relating Hadrurus to Caraboctonus and Hadruroides instead. Whereas Anuroctonus appears to be more closely related to the Vaejovinae, where it was formerly placed, than either taxon is to Hadrurus , we are reluctant to place it back there on the gion to the right); 28, dorsal aspect; 29, ventral aspect. (Total length 3.5 mm) FRANCKE AND SOLEGLAD-THE FAMILY IURIDAE THORELL 245 basis of the hemispermatophore differences noted above. Further study is needed before the sister-group relationships of Anuroctonus can be hypothesized. On the basis of cheliceral morphology, venom glands, and the trichobothrial pattern on the pedipalp chela fixed finger, we hypothesize that Hadrurus is more closely related to Caraboctonus plus Hadruroides than either taxon is to any other Recent scorpion. The relationships of Calchas and Iurus. -These two genera are presently placed each in a monogeneric subfamily in different families: Calchas Birula in the Calchinae (Chac- tidae), and Iurus Thorell in the Iurinae (Vaejovidae). The reason for these placements is obvious: Calchas reportedly has two pairs of lateral eyes, and Iurus has three pairs. Interestingly, however, Vachon (1966:456) diagnosed Iurus as follows: “. . . portant deux yeux lateraux (fig. 19), exceptionnellement trois d’un seul cote (fig. 20)”; yet his figures 19 and 20, based on the holotype of Iurus dekanum (Roewer) [= I. dufoureius (Brulle); Francke, in press ] show four eyes on the left side and three on the right! The single specimen of Calchas we have studied [Turquie, Bilejdik, 23.IV.1971, MNHN-RS 6452, Max VACHON det.] clearly has three lateral eyes on the left side. These observations, coupled with the remarks made in the introduction about the unreliability of this charac- ter, invalidate the previous taxonomic assignments of these two genera. As indicated by Vachon (1971:718, quoted in the introduction), the trichobothrial patterns (Figs. 1-8), and the cheliceral dentition (Figs. 19, 20) of Iurus and Calchas are identical, and we might add unique among Recent scorpions. A clarification is in order regarding cheliceral dentition, for as stated earlier Hadrurus, Hadruroides , and Caraboctonus also have a Figs. 30-33. -Hemispermatophore of Hadruroides charcasus (Karsch): 30, external aspect (dorsal region to the right); 31, dorsal aspect; 32, internal aspect; 33, ventral aspect. (Total length 4.0 mm) 246 THE JOURNAL OF ARACHNOLOGY prominent tooth on the inferior border of the movable finger of the chelicera: these three genera have two subdistal teeth on the dorsal border of the movable finger and have no serrula, while Iurus and Calchas have a single subdistal tooth dorsally and have a serrula (often quite worn down and difficult to see on adult Iurus , but prominent on immatures). The evidence is clear that Iurus and Calchas are more closely related to each other than either one is to any other Recent scorpion. In the past Calchas has played a prominent role in discussions of scorpion phylogeny. Birula (1917) refers to it as the “missing link” between the Buthidae and Chactidae because it has tibial spurs on the third and fourth pairs of legs, as is typical of buthids, whereas other characters indicate chactid affinities. According to Vachon (1971) “Se dans leur ensemble, les Chactidae forment une transition entre Scorpionidae et Buthidae, les Calchinae soulignent avec nettete les affinites entre Chactidae et Vaejovidae” [If Chactidae as a whole form a transition between Scorpionidae and Buthidae, Calchinae clearly underlines the affinities of Chactidae and Vaejovidae]. The fact that Calchas is the only non-buthid Recent scorpion with tibial spurs might lead some to question its hypothesized sister-group relationship with Iurus. However, since tibial spurs are wide- spread among buthids, and are also common on both fossil scorpions and eurypterids (Stormer 1963), their presence on Calchas seems plesiomorphic and uninformative at the level of generalization considered here. Figs. 34-37.-Hemispermatophore of Caraboctonus keyserlingi Pocock: 34, external aspect (dorsal region to the right); 35, dorsal aspect; 36, internal aspect; 37, ventral aspect. (Total length 6.8 mm) FRANCKE AND SOLEGLAD-THE FAMILY IURIDAE THORELL 247 The relationships of lurus + Calchas to Hadrurus + Hadruroides + Caraboctonus .-As indicated by Pocock (1893, quoted as a preface here), lurus and Caraboctonus share certain characters that “point to real kinship between the two.” In light of previous sections, however, their relationships must be considered at a higher level of generaliza- tion: the taxon formed by lurus + Calchas and the taxon formed by Hadrurus + Had- ruroides + Caraboctonus with respect to all other Recent scorpions. The two higher taxa in question share the presence on a prominent tooth on the inferior border of the movable finger of the chelicera. Outgroup comparisons with other members of the two families in which these five genera were formerly placed, as well as comparisons with all Recent scorpions indicate that this is a uniquely derived character shared by them. Thus, we hypothesize they are sister-groups. Among chactoids Caraboctonus, Hadruroides, Hadrurus, and lurus are the only known genera with complex venom glands. Calchas has simple venom glands. Two equally par- simonius hypotheses can be formulated to account for the difference between lurus and Calchas. One hypothesis is that of character reversal in Calchas , i.e., the ancestor of lurus + Calchas had complex venom glands, and after divergence Calchas reverted to simple glands (by neoteny ?). The second hypothesis is that of parallelism, i.e., lurus and Caraboctonus + Hadruroides + Hadrurus independently acquired complex glands. Corrob- oration of the first hypothesis would lead to the conclusion that complex venom glands are a synapomorphy for the two higher taxa under consideration, whereas corroboration of the second hypothesis would refute the case for that synapomorphy. These alternative Figs. 38-41. -He mispermatophore of Anuroctonus phaiodactylus (Wood): 38, external aspect (dor- sal region to the right); 39, dorsal aspect; 40, internal aspect; 41, ventral aspect. (Total length 11.5 mm) 248 THE JOURNAL OF ARACHNOLOGY hypotheses can be tested by analysis of at least three transformation series (Platnick 1977), which are unavailable at this time. The presence of a ventral median row of setaceous tufts in Iurus and Caraboctonus + Hadruroides, a character mentioned by Pocock (1893) as “the hairy clothing of the soles of the feet”, appears to represent a parallelism (by parsimony). It would be most interest- ing, however, to examine first instar specimens of the genera in question because ontoge- netic arguments might shed some light on the polarity of the transformation series (Nelson 1978), and the evidence could be used to expand the cladistic analyses presented here. The hemispermatophore of Iurus (Figs. 53-56) has a distinct truncal flexure but lacks a capsule. Instead, a lightly sclerotized “lobe” develops along the terminal dilation of the vas deferens and the common duct that connects it with the seminal vesicle. Similar “lobes” occur in hemispermatophores of the Superstitioninae (Francke 1981) and may represent an early stage in the evolution of capsules in lamelliform spermatophores. The hemispermatophore of Iurus is similar in some respects to those of Caraboctonus (Figs. 34-37) and Anuroctonus (Figs. 38-41); most striking among these is the absence of a capsule, which as indicated earlier is plesimorphic and thus uninformative. The hemi- spermatophore of Calchas is unknown; Pavolvsky (1924b) illustrated the paraxial organs and these show that a truncal flexure is present as in Iurus. We hereby hypothesize that Calchas + Iurus and Caraboctonus + Hadruroides + Hadrurus form a monophyletic group on the basis of the synapomorphy in their cheli- ceral morphology, i.e., the large tooth on the ventral edge of the movable finger. The Figs. 42-45. -Hemispermatophore of Vaejovis spinigerus (Wood): 42, external aspect (dorsal region to the right); 43, dorsal aspect; 44, internal aspect; 45, ventral aspect. (Total length 7.0 mm) FRANCKE AND SOLEGLAD-THE FAMILY IURIDAE THORELL 249 taxonomic chaos prevalent among chactoids prevents us from carrying the cladistic anal- ysis further at this time. The sister group of the monophyletic taxon proposed above remains unknown, and once it is recognized further characters will become available for analysis. Likewise, as more monophyletic groupings within the chactoids are identified, their relationships will be easier to reconstruct. The phylogenetic relationships hypothesized in this section are diagrammatically illus- trated by the cladogram in Fig. 57. CLASSIFICATION AND TAXONOMY The phylogenetic information contained in the cladogram (Fig. 57) is also expressed by the following sequenced classification: Family Iuridae Thorell, 1 876, new rank Subfamily Iurinae Thorell, 1876 Iurus Thorell, 1876 Calchas Birula, 1 899 Subfamily Caraboctoninae Kraepelin, 1905 Tribe Caraboctonini, new Caraboctonus Pocock, 1 893 Hadruroides Pocock, 1 893 Tribe Hadrurini, new Hadrurus Thorell, 1876 The taxonomic changes resulting from this classification follow. KEY TO SCORPION FAMILIES 1. Sternum subtriangular. Cheliceral movable finger with dorsal tine longer than ventral tine; with two basal teeth. Pedipalp femur with 10 or more trichobothria, of which 4-5 are on internal aspect. Pedipalp tibia without ventral trichobothria .... Buthidae Sternum not subtriangular, but pentagonal or slit-like. Cheliceral movable finger with dorsal tine shorter than ventral tine; with only one basal tooth. Pedipalp femur with 9 or fewer trichobothria, of which only one on internal aspect. Pedipalp tibia with one or more ventral trichobothria 2 2. Gnathocoxa broadly expanded anteriorly. Pedipalp femur with nine trichobothria, of which four are on dorsal aspect. Ventral trichobothria of tibia along ventrointernal keel Chaerilidae Gnathocoxa not broadly expanded anteriorly, but tapering gradually. Pedipalp femur with 3-4 trichobothria, of which only one on dorsal aspect. Ventral trichobothria of tibia along ventroexternal keel 3 3. Retrolateral pedal spurs absent. Female ovariuterus with conspicuous diverticula . . 4 Retrolateral pedal spurs present. Female ovariuterus without diverticula 5 250 THE JOURNAL OF ARACHNOLOGY Figs. 46-49. -Hemispermatophore of Uroctonus apacheanus Gertsch and Soleglad: 46, external aspect (dorsal region to the right); 47, dorsal aspect; 48, internal aspect; 49, ventral aspect. (Total length 5.0 mm) 4. Subaculear tubercle present Diplocentridae Subaculear tubercle absent Scorpionidae 5. Sternum reduced to transverse, slit-like sclerite Bothriuridae Sternum not reduced to transverse, slit-like sclerite, but well developed and sub- pentagonal 6 6. Cheliceral movable finger with one well developed tooth on ventral margin. Venom glands complex, folded (except simple, unfolded in Calchas, which also has tibial spurs on legs III and IV) Iuridae Cheliceral movable finger without well developed tooth on ventral margin (several small denticles and/or tubercles may be present). Venom glands simple, unfolded “Chactoids” (Chactidae + Vaejovidae) FAMILY IURIDAE THORELL Type genus —Iurus Thorell, 1876. Included taxa.— Iurinae Thorell, 1876, and Caraboctoninae Kraepelin, 1905. FRANCKE AND SOLEGLAD-THE FAMILY IURIDAE THORELL 251 Diagnosis.— (1) Sternum pentagonal. Cheliceral movable finger (2) with dorsal distal tooth shorter than ventral distal tooth, and (3) with prominent ventral basal tooth. Pedipalp femur (4) with three trichobothria. Male reproductive system (5) without well developed, prominent accessory glands; (6) spermatophore lamelliform. Female ovariu- terus (7) with four pairs of symmetrical transverse anastomoses, (8) lacking diverticula; (9) ova with little or no yolk. Venom glands (10) with several folds. Supraneural lym- phatic gland (11) extends the length of the mesosoma, and two saclike lymphoid organs arise as diverticula of the diaphragm. Comparisons.— The family Iuridae differs from the Buthidae in characters 1, 2, 3, 4, 5, 6, 7, 8, 9, and 1 1 above. Differs from Chaerilidae in that the gnathocoxa of leg I are not expanded anteriorly, and in characters 3, 4, 9, and 10. The Scorpioniodea are easily separated by characters 3 and 8; the Bothriuridae by characters 1,3, and 9; and the chactoids by characters 3 and 10. Subfamily Iurinae Thorell Type genus —lurus Thorell, 1876. Figs. 50-52. -Hemispermatophore of Paruroctonus utahensis (Williams): 50, external aspect (dorsal region to the right); 51, dorsal aspect; 52, ventroexternal aspect; 52, ventroexternal aspect. (Total length 7.0 mm) 252 THE JOURNAL OF ARACHNOLOGY Included taxa .—Iurus Thorell (Aegean Sea islands, Turkey, Greece), Calchas Birula, 1899 (Turkey; Georgian S.S.R., Russia). Diagnosis.— Cheliceral movable finger with (1) a single subdistal tooth dorsally, and (2) a serrula ventrally (Figs. 19, 20). Trichobothrial pattern (Figs. 1-8, 12, 16): (3) tibia with a single trichobothrium ventrally; (4) internal aspect of chela fixed finger with one trichobothrium on distal one-half. Comparisons.— Differs from the Caraboctoninae in characters 1-4 above (see Key to subordinate taxa of Iuridae). Remarks.— The two genera in this subfamily are easily separated by differences in pedipalp finger dentition, armature of the tibia of legs III and IV, setation on ventral aspect of tarsus, and shape of the stigmata (see Key to subordinate taxa). Subfamily Caraboctoninae Kraepelin Type genus —Caraboctonus Pocock, 1893. Included taxa.-Caraboctonini, new (western South America), Hadrurini, new (western North America). Diagnosis.— Cheliceral movable finger with (1) two subdistal teeth dorsally, and (2) without serruly ventrally (Figs. 17, 18). Trichobothrial pattern (Figs. 9, 10, 13, 14): (3) tibia with two or more trichobothria ventrally; (4) internal aspect of chela fixed finger without trichobothria on distal one-half. Figs. 53-56.-Hemispermatophore of Iurus sp.: 5 3, external aspect (dorsal region to the right); 54, dorsal aspect; 55, internal aspect; 56, ventral aspect. (Total length 14.0 mm) FRANCKE AND SOLEGLAD-THE FAMILY IURIDAE THORELL 25 3 Comparisons.— Differs from the Iurinae in characters 1-4 above (see Key to subordi- nate taxa of Iuridae). Remarks.— The two tribes in this subfamily are easily separated by differences in pedipalp finger dentition, trichobothrial patterns, and setation on ventral aspect of tarsus (see Key to subordinate taxa). Additional differences include the hemispermatophores (Figs. 23-37), development of genital papillae on males, structure of pedal spurs, and development of unguicular claw (see respective diagnoses). Tribe Caraboctonini, new Type genus.— Caraboctonus Pocock, 1893. Included taxa.— Caraboctonus Pocock (central Chile), Hadruroides Pocock, 1893 (Peru, Ecuador). Diagnosis. -Pedipalp chela fingers with (1) 6-7 oblique rows of principal denticles. Trichobothrial pattern (Figs. 9, 13): tibia with (2) 15 trichobothria on external aspect, and (3) two on ventral aspect; palm of chela (4) without dorsal trichobothria, with (5) five trichobothria on ventral aspect. Legs with (6) pedal spurs simple; tarsus with (7) setaceous tufts ventrally, and (8) unguicular claw poorly developed. Males with (9) genital papillae well developed. Comparisons.— Differs from Hadrurini in characters 1-9 above (see Key to subordinate taxa, and diagnosis of Hadrurini). Remarks.— The two genera in this tribe can be separated by differences in pedipalp finger dentition, and shape of the sternum (see Key to subordinate taxa). Fig. 57.-Cladogram illustrating the hypothesized phylogenetic relationships within the family Iuridae Thorell. 254 THE JOURNAL OF ARACHNOLOGY Tribe Hadrurini, new Type genus —Hadrurus Thorell, 1876. Monotypic. Diagnosis.— Pe dipalp finger with (1) 9-10 oblique rows of principal denticles. Tricho- bothrial pattern (Figs. 10, 14): tibia with (2) over 50 trichobothria on external aspect, and (3) over 30 on ventral aspect; palm of chela with (4) two dorsal trichobothria, and (5) over 12 on ventral aspect. Legs with (6) pedal spurs pectinate; tarsus (7) without setaceous tufts ventrally, and with (8) unguicular claw strongly developed. Males (9) without genital papillae. Comparisons.— Differs from the Caraboctonini in characters 1-9 above (see Key to subordinate taxa, and diagnosis of Caraboctonini). KEY TO SUBORDINATE TAXA OF IURIDAE 1. Cheliceral movable finger with one subdistal tooth dorsally, with serrula ventrally; pedipalp tibia with only one trichobothrium on ventral aspect; pedipalp chela fixed finger with one trichobothrium on distal one-half of internal aspect Iurinae ... 2 Cheliceral movable finger with two subdistal teeth dorsally, without serrula; pedipalp tibia with more than one trichobothrium on ventral aspect; pedipalp chela fixed finger without trichobothria on distal one-half of internal aspect Caraboctoninae ... 3 2. Pedipalp chela fingers with 6-7 oblique rows of principal denticles; tibial spurs present on third and fourth pairs of legs; tarsus with two submedian rows of setae ventrally, without median row of setaceous tufts; stigmata small, oval Calchas Pedipalp chela fingers with 14-15 oblique rows of principal denticles; tibial spurs absent; tarsus with two submedian rows of setae ventrally, and with median row of setaceous tufts; stigmata long, slit-like Iurus 3. Pedipalp chela fingers with 9-10 oblique rows of principal denticles; tarsus with two submedian rows of setae ventrally, and with median row of short spines; pedipalp tibia with more than two (over 30 usually) trichobothria on ventral aspect, chela with more than four (about 13 to 27) on ventral aspect Hadrurini . . . Hadrurus Pedipalp chela fingers with 6-7 oblique rows of principal denticles; tarsus with two sub median rows of setae ventrally, and with median row of setaceous tufts; pedipalp with only two trichobothria on ventral aspect, chela with four trichobothria on ventral aspect Caraboctonini ... 4 4. Pedipalp chela fingers on adults and subadults with internal and external super- numerary granules flanking the oblique rows of principal denticles; sternum as long as wide, with a deep longitudinal furrow Hadruroides Pedipalp chela fingers on adults and subadults without internal and external super- numerary granules flanking the oblique rows of principal denticles; sternum wider than long, without a deep longitudinal furrow (at most with a deep pit posteriorly) .... Caraboctonus FRANCKE AND SOLEGLAD-THE FAMILY IURIDAE THORELL 255 ZOOGEOGRAPHY OF THE IURIDAE The hypothesized phylogenetic relationships within the Iuridae are quite simple, as shown in the cladogram (Fig. 57). The basal dichotomy gives rise to the subfamilies Iurinae and Caraboctoninae: in the former a single subsequent dichotomy gives rise to the two genera of iurines; whereas on the latter the following dichotomy gives rise to the two tribes of caraboctonines, and a subsequent dichotomy in the Caraboctonini produced the two genera it contains. In addition, Soleglad (1976) recognized two species groups in Hadrurus : the aztecus group with two species in south-central Mexico (Oaxaca, Puebla, and Guerrero), and the hirsutus group with six species in Baja California and the south- western United States. No futher subdivisions have been proposed for Hadruroides, the only other polytypic genus in the subfamily. The geographical distribution of the family Iuridae might appear unusual to some, with Calchas in the Caucasus, Iurus with two species in the Aegean region (Francke, in press), Hadrurus in western North America, and Hadruroides and Caraboctonus in west- ern South America (Fig. 58). This disjunct distribution nonetheless belongs to a biogeo- graphic track, the “Tethys geosyncline,” shared with many other organisms, among which the Malvaceae Malopeae, with allied genera in Chile, Peru, Mexico, the Mediterranean, Hungary and the Balkans, is a good example (Croizat 1958, chapter II). The New World Caraboctoninae form part of the well known Southwestern Peru- Galapagos-Mexico track (Croizat 1958, chapter VIII). This track often involves the Revil- lagigedo Islands (off the Pacific coast of Mexico), from which only a Vaejovis sp. has been reported (Williams, 1980). The concepts of vicariance biogeography predict, however, the presence (now or in the past) of a caraboctonine on Revillagigedo, and it would be extremely interesting to collect more scorpions on those islands to test that prediction. Geotectonic events possibly responsible for the vicariance patterns observed in iurids are first the opening of the North Atlantic during Jurassic times (Hallam 1971, Sclater and Tapscott 1979), isolating the two subfamilies on either side of that ocean. Secondly, the decoupling of the North American and South American plates, which formed a prominent role in the formation of the Caribbean region, during late Mesozoic to early Tertiary times (Rosen 1976), may have isolated Hadrurus in western North America and 256 THE JOURNAL OF ARACHNOLOGY the Caraboctonini in western South America. By late Eocene to early Oligocene times, the Galapagos spreading center became active (Rosen 1976), leading eventually to the isolation of Hadruroides maculatus galapagoensis Maury on those islands. Likewise, the opening of the Gulf of California during the Pliocene (Moore and Buffington 1968) may have resulted in the split of the aztecus group of Hadrurus in south-central Mexico and the hirsutus group in Baja California and the southwestern United States. The break between Iurus and Calchas is probably related to tectonism involving the Turkish plate and the Anatolian fault sometime during the Tertiary. More precise dating should be possible as the geophysical history of that region becomes better known. ACKNOWLEDGMENTS For the loan of specimens we thank Dr. 0. Elter, Museo ed Istituto di Zoologia Sistematica della Universita, Torino; Dr. J. Gruber, Naturhistorisches Museum Wien, Vienna; Dr. E. A. Maury, Museo Argentino de Ciencias Naturales, Buenos Aires; Dr. N. I. Platnick, American Museum of Natural History, New York; Prof. M. Vachon, Museum National d’Histoire Naturelle, Paris; and the late Dr. E. N. K. Waering, Florida. We are most gratefull to Dr. E. A. Maury, Dr. S. C. Williams, Dr. N. I. Platnick, and Mr. W. D. Sissom for their comments and suggestions on the manuscript. LITERATURE CITED Birula, A. A. B. 1899. Miscellanea Scorpiologica. Ezheg. Zool. Muz. Imp. Akad. Nauk, 4:14-19 (in Russian). Birula, A. A. B. 1917. Fauna of Russia and Adjacent countries. Arachnoidea, Scorpions. 1(1): 1-224. (English translation by Israeli Program for Scientific Translations, Jerusalem 1965, 154 pp.). Croizat, L. 1958. Panbiogeography. Vol. 1 -The New World. Published by the author, Caracas, Venezuela, 1018 pp. Curcic, B. P. M. 1972. Considerations upon the geographic distribution and origin of some populations in the genus Euscorpius Thorell. Rapp. Comm. int. Mer Medit., 21:83-88. Francke, O. F. 1976. Redescription of Parascorpiops montana Banks. Entomol. News, 87:75-85. Francke, O. F. 1979. Spermatophores of some North American scorpions. J. Arachnol., 7:19-32. Francke, O. F. 1981. Studies of the scorpion subfamilies Superstitioninae and Typhlochactinae, with description of a new genus (Scorpiones, Chactoidea). Assoc. Mexican Cave Stud., Bull. 7: Francke, O. F. in press. Taxonomic and zoogeographic observations on Iurus Thorell (Scorpiones, Iuridae). Bull. British Arachnol. Soc. Francke, O. F. and M. E. Soleglad. 1980. Two new Hadruroides Pocock, from Peru. Occas. Papers Museum, Texas Tech University, 69:1-13. Gertsch, W. J. and M. E. Soleglad 1972. Studies of North American scorpions of the genera Uroctonus and Vejovis. Bull. Amer. Mus. Nat. Hist., 148:549-608. Gonzales, M. A. 1977. Rectification del caracter “ojos laterales” en varios generos de la familia Chactidae (Scorpionida) en Venezuela. Acta Biol. Venezuela, 9:303-315. Hallam, A. 1971. Mesozoic geology and the opening of the North Atlantic. J. Geol., 79:129-157. Jespersen, A. and R. Hartwick. 1973. Fine structure of spermiogenesis in scorpions from the family Vejovidae. J. Ultrastruct. Res., 45:366-383. Koch, L. E. 1977. The taxonomy, geographic distribution and evolutionary radiation of Australo- Papuan scorpions. Rec. Western Australian Mus., 5(2) :8 1-367. Kraepelin, K. 1905. Die geographische Verbreitung der Scorpione. Zool. Jahrb., 22:321-364. Lamoral, B. H. 1980. A reappraisal of suprageneric classification of recent scorpions and of their zoogeography. 8th. Int. Cong. Arachnol., Vienna 1980, pp. 439-444. Maury, E. A. 1973. Las tricobotrias y su importancia en la sistematica del genero Brachistosternus Pocock, 1894. Physis, sec. C, 32:247-254. FRANCKE AND SOLEGLAD-THE FAMILY IURIDAE THORELL 257 Maury, E. A. 1975. Escorpiones y escorpionismo en el Peru. IV: Revision del genero Hadruroides Pocock, 1893. Rev. Peruana Entomol., 17:9-21. Maury, E. A. 1976. Escorpiones y escorpionismo en el Peru. V: Orobothriurus, un nuevo genero de escorpiones altoandino. Rev. Peruana Entomol., 18:14-25. Moore, D. G. and E. C. Buffington 1968. Transform faulting and growth of the Gulf of California since Late Pliocene. Science, 161:1238-1241. Nelson, G. 1978. Ontogeny, phylogeny, and the biogenetic law. Syst. Zool., 27:324-345. Pavlovsky, E. N. 1924a. Studies on the organization and development of scorpions. Quart. J. Micr. Sci., 68:615-640. Pavlovsky, E. N. 1924b. On the morphology of the male genital apparatus in scorpions. Trav. Soc. Nat. Leningrad, 53:76-86. Platnick, N. I. 1977. Parallelism in phylogeny reconstruction. Syst. Zool., 26:93-96. Pocock, R. I. 1893. Notes on the classification of scorpions, followed by some observations upon synonymy, with descriptions of new genera and species. Ann. Mag. Nat. Hist., ser. 6, 12:303-330. Probst, P. J. 1972. Zur Fortpflanzungsbiologie and zur Entwicklung der Giftdriisen beim Skorpion Isometrus maculatus (DeGeer, 1778). Acta Tropica, 29:1-87. Rosen, D. E. 1976. A vicariance model of Caribbean biogeography. Syst. Zool., 24:431-464. Sclater, J. G. and C. Tapscott. 1979. The history of the Atlantic. Sci. Amer., 240(6):156-174. Soleglad, M. E. 1973. Uroctonoides fractus a synonymy. Pan-Pacific Entomol., 49:60. Soleglad, M. E. 1976a. A revision of the scorpion subfamily Megacorminae (Scorpionida: Chactidae). Wasmann J. Biol., 34:251-303. Soleglad, M. E. 1976b. The taxonomy of the genus Hadrurus based on chela trichobothria. J. Arach- nol., 3:113-134. Stahnke, H. L. 1974. Revision and keys to the higher categories of Vejovidae. J. Arachnol., 1:107-141. Stormer, L. 1963. Gigantoscorpio willsi, a new scorpion from the Lower Carboniferous of Scotland and its associated preying microorganisms. Skrifter utgett Norse Videnskaps-Akademi Oslo. I. Mat.-Naturv. Klasse. NY ser., No. 8, 171 pp. Thorell, T. 1876. Etudes scorpiologiques. Atti Soc. Italiana Sci. Nat., Milano, 19:75-272. Vachon, M. 1952. Etudes sur les scorpions. Inst. Pasteur d’Algerie, Alger, 482 pp. Vachon, M. 1963. De l’utilite, en systematique, d’une nomenclature des dents des cheliceres chez les scorpions. Bull. Mus. Nat. Hist. Nat., Paris, 2e. ser., 35(2) : 1 6 1-1 66 . Vachon, M. 1966. A propos de la synonymie de deux genres de scorpions: Chaerilomma Roewer, 1943 (Chactidae) et Iurus Thorell, 1877 (Vejovidae). Senck. biol., 47(6):453-461 . Vachon, M. 1971. Remarques sur le scorpions caucasien Calchas nordmanni Birula. Rev. Entomol. URSS, 50:712-718 (in Russian). Vachon, M. 1972. Sur l’etablissement d’une nomenclature trichobothrial uniforme convenant a l’en- semble des scorpions (Arachnides) et l’existence de trois types distincts de trichobothriotaxie. C. R. Acad. Sci. Paris, ser. D, 275:2001-2004. Vachon, M. 1974. Etude des caracteres utilises pour classer les families et les genres de scorpions (Arachnides). 1. La trichobothriotaxie en Arachnologie. Sigles trichobothriaux et types de tricho- bothriotaxie chez les scorpions. Bull. Mus. Nat. Hist. Nat., Paris, 3e. ser., 140 (Zool. 104):857-958. Vachon, M.. 1975. Recherches sur les scorpions appartenant ou deposes au Museum d’Historie nat- urelle de Geneve. I. Contribution a une meilleure connaisance des especes et des sous-especes de scorpions due genr e Euscorpius Thorell, 1876. Rev. Suisse Zool., 82:629-645. Valle, A. 1975. Considerazioni intorno alle sottospecie di Euscorpius carpathicus (L.). Ateneo Par- mense, act nat., 11:209-234. Williams, S. C. 1980. Scorpions of Baja California, Mexico, and adjacent islands. Occas. Papers, California Acad. Sci., No. 135, 127 pp. Manuscript received March 1980, revised July 1980. Millidge, A. F. 1981. The erigonine spiders of North America. Part 4. The genus Disembolus Chamber- lin and Ivie (Araneae, Linyphiidae). J. Arachnol., 9:259-284. THE ERIGONINE SPIDERS OF NORTH AMERICA. PART 4. THE GENUS DISEMBOLUS CHAMBERLIN AND IVIE (ARANEAE: LINYPHnDAE) A. F. Millidge Little Farthing, Upper Westhill Road, Lyme Regis, Dorset DT7 3ER, England ABSTRACT The erigonine genus Disembolus Chamberlin and Ivie, which appears to be endemic to North America, has been revised. Cochlembolus sacerdotalis Crosby and Bishop, Soudinus corneliae Cham- berlin and Ivie, Tapinocyba alpha Chamberlin, Tapinocyba kesimba Chamberlin and Tapinocyba (?) phana Chamberlin have been transferred to Disembolus , while Disembolus apache Chamberlin and (in the absence of the type) Disembolus zygethus Chamberlin have been excluded from the genus. The somatic characters of the Disembolus species are very similar to those of the genera Spirembolus Chamberlin and Scotinotylus Simon, and Disembolus can be adequately defined and differentiated only on the structure of the male palpal organs and of the female epigyna. Synapomorphic genitalic characters for the genus have been identified. The genus now contains 22 species, including the following 16 new taxa: Disembolus amoenus, D. anguineus, D. beta, D. concinnus, D. convolutus, D. galeatus, D. hyalinus, D. implexus, D. implicatus, D. lacteus, D. lacunatus, D. procerus, D. sinuosus, D. solanus, D. torquatus and D. vicinus. Descriptions, diagnoses and distribution maps are given for all the species. INTRODUCTION The genus Disembolus was erected by Chamberlin and Ivie (1933) for the single species D. stridulans Chamberlin and Ivie, and was redescribed some years later (Cham- berlin and Ivie 1945). Several species were added to the genus by Chamberlin (1948). Examination of material supplied by the American Museum of Natural History (AMNH), the Museum of Comparative Zoology, Harvard University (MCZ) and the Canadian National Collection, Ottawa (CNC), has disclosed a number of additional species, and has made possible a more detailed description of the genus. Genus Disembolus Chamberlin and Ivie Disembolus Chamberlin and Ivie 1933:20; 1945:225 (type species: Disembolus stridulans Chamberlin and Ivie 1933, by original designation). The members of this genus are small spiders, none of which exceeds 2 mm in total length. The female carapace is moderately raised behind the eyes (e.g. Fig. 69). The male. 260 THE JOURNAL OF ARACHNOLOGY carapace usually bears a lobe anteriorly, with holes and sulci behind the lateral eyes (Figs. 57-61); in most species the lobe does not carry the eyes. In D. stridulans, D. procerus, new species, and D. kesimbus (Chamberlin) the holes and sulci are absent, and the posterior median eyes are carried on the summit of the lobe (Figs. 12, 55). The posterior median eyes of the female are in every case more widely spaced from one another than from the laterals. All the species have files on the lateral margins of the chelicerae in both Map 1.- Distribution of Disembolus species in North America: A. D. alpha, D. anguineus, D. hyalinus, D. sihuosus, D. corneliae and D. sacerdotalis B. D. stridulans, D. procerus, D. torquatus, D. galeatus and D. solanus C. D. implicatus, D. beta, D. implexus, D. vicinus, D. lacteus andD. convolutus D. D. concinnus, D. kesimbus, D. amoenus, D. phanus and D. lacunatus MILLIDGE-GENUS DISEMBOLUS 261 sexes. The abdomen is without scuta and is more or less unicolorous; only a few species have stridulatory files on the epigastric plates. The legs are relatively short and stout, with a value for tibia I 1/d (female) of 5-7. In all species the tibial spines are 2221 in the female, but are usually reduced in number in the male. Metatarsi I-III have a dorsal trichobothrium, which is absent on metatarsus IV; the value of Tml (female) is 0.35-0.55. The male palpal tibia is produced dorsally into a short apophysis which sometimes has a pointed hook distally (e.g. Fig. 49); the tibia always bears one stout spine dorsally (less developed in D. sinuosus, new species), and usually two trichobothria, but occasionally only one. The female palpal tibia always has two trichobothria dorsally. The characters enumerated above are reminiscent of those given for the genera Spir- embolus (Millidge 1980) and Scotinotylus (Millidge, 1981), and do not serve to differentiate Disembolus from related genera. The genus can be properly defined only on the structure of the genitalia. The paracymbium of the male palp is U-shaped with the distal arm shorter and pointed (e.g. Fig. 15). The tegulum in most instances is truncated anteriorly (e.g. Fig. 26), but in the type species it has a pronounced anterior projection (Fig. 9). The embolic division (ED) of the palpal organ has an elongated central section lying between the tailpiece and the embolus proper (Figs. 1 , 2, 3). The embolus has basically a spiral form, but distally it becomes convoluted to a greater or lesser degree; the terminal part of the embolus is always retroverted, with the tip in many instances lying more or less within a coil of the embolus (e.g. Figs. 2, 3). In D. stridulans and D. procerus the embolus has a very tangled appearance (Figs. 3, 9, 13, 15), but in other species the convolutions are developed to a lesser degree (e.g. Figs. 2, 33). The central section of the ED is equipped with what seems to be a reinforcing member (R, Figs. 1, 2, 3), located just posterior to the beginning of the embolus proper, and which sometimes gives the false impression of being the basal part of the embolus. The tailpiece of the ED is screw-like, with the distal part often somewhat lengthened (e.g. Fig. 1); this distal part is not particularly long in D. stridulans , while D. kesimbus (Chamberlin), which has a generally less highly developed ED, has a relatively short tailpiece (Fig. 17). The suprategular apophysis (SA) comprises (i) a weakly sclerotized part which is tusk or spade shaped and runs ventrad from the end of the suprategulum (TK, Figs. 4, 5, 6, 7, 8, 16, 32); (ii) a fairly robust membraneous part of variable form which runs anteriad from the end of the suprategulum on the mesal side of the tusk-shaped part (M, Figs. 4, 5, 6, 7, 8); and (iii) an auxiliary membrane, thin and transparent, which extends between parts (i) and (ii) (AM, Figs. 4, 5, 6, 7, 8). The stalk (S, Fig. 5), which carries the seminal duct to the ED, is more or less continuous with the membraneous part (ii). The ED of Disembolus is of the same general pattern as in Spirembolus and Scotinotylus , but it differs from these in two significant respects: firstly, the embolus is distally more complex in form, and has the tip retroverted, and secondly, the central section of the ED is elongated. The SA is most similar to that of Scotinotylus , but differs in the shape of the membraneous part and particularly in the presence of the auxiliary membrane. The forms of the ED and of the SA can be regarded as derived palpal characters common to all members of the genus; these synapomorphies support the hypothesis that the genus as constituted here is a monophyletic group. The epigyna of the Disembolus species have posteriorly a somewhat convex, trans- lucent plate, roughly trapezoidal or ellipsoidal in shape (e.g. Figs. 74, 84), which often has a glassy, lens-like appearance; through this plate the outlines of the internal ducts, etc. are faintly visible. The spermathecae can usually be seen through the epigynal integu- ment. In several of the species there is a darker colored ridge or mantle anterior to the 262 THE JOURNAL OF ARACHNOLOGY Table 1. -Partial key to Disembolus species: females. A decision on species identity should be made only after reference to the species descriptions and diagnoses. 1. Posterior plate of epigynum distinctly notched anteriorly a. epigynum as Fig. 72, with 2 small clear markings within plate D. kesimbus b. epigynum as Fig. 7 3 D. torquatus 2. Epigynum with dark colored knob or ridge on anterior margin a. epigynum as Fig. 75; spermathecae close together D. anguine us b. epigynum as Fig. 74; spermathecae further apart D. convolutus 3. Epigynum with dark colored ridge or mantle just anterior to the posterior plate (e.g. Fig. 79) (note: the epigynum must be examined vertically - when viewed somewhat from behind, the mantle is much less clear) a. mantle roughly triangular in shape i. mantle small (Figs. 76, 78) D. amoenus, D. sinuosus (for separation, see species descriptions) ii. mantle larger (Fig. 79) D. concinnus b. mantle roughly trapezoidal in shape (Figs. 81, 82) D. comeliae 4. Epigynum as Fig. 80, with posterior plate notably convex and glassy in appearance, and with dark colored bar anterior to plate D. hyalinus 5. Epigynum not of the form given in 1-4 (Figs. 83-91) a. clear banana-shaped markings within the plate (Fig. 83); a tiny spider (total length 1.0-1. 2 mm) D. alpha b. markings within plate faint or absent i. plate milky white in color (Fig. 87) D. lacteus ii. spermathecae rather closely spaced (Fig. 88) D. vicinus iii. spermathecae rather widely spaced (Figs. 89-91) D. procerus, D. galeatus, D. solanus (for separation, see species descriptions) iv. spermathecae moderately spaced (Figs. 84, 85, 86) D. implicatus, D. phanus (for separation, see species descriptions) plate (e.g. Fig. 81). The openings of the spermathecal ducts seem to lie towards the posterior margin of the plate (e.g. Fig. 99), but there is some uncertainty as these are very difficult to see. The ducts arise on the dorsal side of the spermathecae, and follow a fairly simple curved pathway to the external openings. Since the ducts are transparent and practically unpigmented, it must be accepted that the figures given for the internal genitalia (Figs. 92-108) may not be accurate in every detail. The epigyna can be recognized in most cases by the presence of the translucent plate posteriorly. This feature, coupled with the simple arrangement of the spermathecal ducts and the position of the genital openings, probably represent a synapomorphic character for the genus. It is difficult to visualize how a Disembolus male can transfer sperm efficiently into the spermathecal duct of the female. The tip of the embolus is retroverted, often penetrating backwards well into the embolic coil; this tip does not uncoil or straighten when the palp is expanded. To make the situation even more difficult, the female has an apparently smooth and glossy epigynal plate which has no grooves or holds to guide the MILLIDGE-GENUS DISEMBOLUS 263 Table 2.-Partial key to Disembolus species: males. A decision on species identity should be made only after reference to the species descriptions and diagnoses. 1. Carapace with no lobe, but with a tiny hole behind the lateral eyes (Fig. 56); a tiny spider (total length 1.0- 1.2 mm) D. alpha 2. Carapace elevated anteriorly into a lobe which has no holes or sulci behind the lateral eyes (Figs. 12,14,55) D. stridulans, D. procerus, D. kesimbus (for separation, see species descrip- tions) 3. Carapace with a lobe which has a hole and sulcus on each side (e.g. Fig. 60) a. palpal tibial apophysis, viewed laterally, with a forward-directed point distally (e.g. Figs. 16, 49) i. tibial apophysis as Figs. 49, 50 D. phanus, D. lacunatus (for separation, see species descriptions) ii. tibial apophysis as Figs. 39, 41, 47 D. implicatus, D. implexus, D. convolutus (for separation, see species de- scriptions) iii. tibial apophysis as Fig. 18; SA with upward curving point distally (Figs. 18, 19) D. torquatus iv. tibial apophysis long, with small hook distally (Fig. 31); a relatively large species (total length 1. 9-2.0 mm) D. sacerdotalis b. palpal tibial apophysis, viewed laterally, not pointed distally (e.g. Figs. 24, 32) i. tibial apophysis fairly short and turned over at distal end (Figs. 24, 26) D. beta, D. sinuosus (for separation, see species descriptions) ii. tibial apophysis longer and not turned over at distal end (Fig. 32) D. anguineus embolus to the very inconspicuous genital pores. Does the male perhaps never achieve insertion of the embolus, but instead eject the sperm on to the surface of the epigynum around the genital pores? It would be interesting to examine, with live specimens of these tiny spiders, how the engagement of the palp with the epigynum takes place. Species transferred into Disembolus— On the basis of the genital structures, the follow- ing species have been transferred into Disembolus : Cochlembolus sacerdotalis Crosby and Bishop 1933 Soudinus corneliae Chamberlin and Ivie 1944 Tapinocyba alpha Chamberlin 1948 Tapinocyba kesimba Chamberlin 1948 Tapinocyba (?) phana Chamberlin 1948 The figure given by Chamberlin (1948) for the epigynum of Oedothorax cascadeus in- dicates that this species may possibly be a Disembolus', the single specimen described cannot however be found, and hence its identity must remain uncertain. A similar situation exists for Tapinocyba (?) pontis Chamberlin 1948. Misplaced species —Disembolus apache Chamberlin 1948 has been transferred to Scotinotylus (Millidge 1981). The female of Disembolus zygethus Chamberlin 1948 has not been located, and no opinion can be given on this species. The putative female of Disembolus stridulans (Chamberlin and Ivie 1945) is Scotinotylus sanctus (Crosby) (Millidge 1981). Distribution and Natural History.— The genus Disembolus appears to be endemic to North America. The species are distributed throughout the United States, but the majority have been found only in the western half. There is only one record for Canada 264 THE JOURNAL OF ARACHNOLOGY and none for Mexico. No valid conclusions on distribution can be drawn, however, since none of the species is represented by more than a few specimens. Little is known on the natural history of the Disembolus species. A few have been recorded from under stones or in vegetable detritus, some at relatively high altitudes. Like most erigonines, they presumably live almost exclusively at or below ground level. The sparsity of material in the collections may indicate that most of the species have a limited distribution or live in specialized or unusual habitats. It should be remembered, however, that some tiny erigonines may give the impression of extreme rarity when hand collecting is employed, but are taken in large numbers when pitfall trapping is carried out in the same area. Species.-The genus as now defined contains 22 known species. More than half of these are described on one sex only, and it is possible that the male and female of a single species are described under separate names in a few instances. The considerable number of species represented in the comparatively few vials of material received from all sources Figs. 1-8.— 1, D. torquatus , embolic division; 2, D. anguineus, embolic division; 3, D. stridulans, embolic division; 4, D. stridulans, male palp, meso-ventral, embolic division removed; 5,D. torquatus, suprategular apophysis, mesal; 6, D. torquatus, suprategular apophysis, ectal; 7, D. anguineus, suprategular apophysis, ectal; 8, D. anguineus, suprategular apophysis, dorso-ectal. Abbreviations: AM, suprategular apophysis, auxiliary membrane; E, embolus; M, suprategular apophysis, membraneous part; R, reinforcing member; S, stalk; T, tegulum; TK, suprategular apophsis, tusk-like part, TP, tailpiece (Scale lines 0.1 mm) MILLIDGE-GENUS DISEMBOLUS 265 makes it probable that several more species remain to be discovered. The species described in this paper are as follows: Disembolus stridulans Chamberlin and Ivie D. procerus , new species D. galeatus , new species D. solanus, new species D. kesimbus (Chamberlin and Ivie) D. torquatus, new species D. alpha (Chamberlin) D. beta , new species D. sinuosus, new species D. sacerdotalis (Bishop and Crosby) D. anguineus, new species D. implicatus, new species D. implexus, new species D. convolutus , new species D. phanus (Chamberlin) D. lacunatus, new species D. lacteus, new species D. vicinus, new species D. amoenus, new species D. concinnus , new species D. corneliae (Chamberlin and Ivie) D. hyalinus , new species Keys to the species.— Partial keys to the Disembolus species are presented in Tables 1 and 2. Apart from some differences in size, the females are all very similar to one another, and their determination relies almost entirely on the form of the epigynum. The males show somewhat greater differences, not only in the palps but also in the form of the carapace lobes. The keys, particularly for the females, should not be used uncritically, and with both sexes a final diagnosis should be made only after reference to the species descriptions and diagnoses. In doubtful cases, bearing in mind the likelihood that new species will be discovered, comparisons with the type or paratypes will be desirable. Descriptions of species.— The descriptions follow the order given in the list of species above. Figures of the male palps are of the right hand palp. Disembolus stridulans Chamberlin and Ivie Figures 3, 4, 9, 10, 11, 12, 13; Map IB Disembolus stridulans Chamberlin and Ivie 1933:21; 1945:226 (mal e,not female); Roewer 1942:664; Bonnet 1956:1516 Type.— Male holotype from Raft River Mts., south fork of Raft River, 8 mi. south of Lynn, Utah, September 6, 1932 (W. Ivie); in AMNH. Paratypes examined. Description.— Only the male is known. Total length: male 1.35-1.40 mm. Carapace: length: male 0.65 mm. Brown to orange-brown, with blackish markings and margins. The carapace is raised anteriorly into a lobe which projects over the clypeus (Fig. 12); there are no holes or sulci. Chelicerae: with clear stridulatory file (Fig. 11). Abdomen: grey to black; epigastric plates smooth. Sternum: practically black. Legs: brown to yellow-brown. 266 THE JOURNAL OF ARACHNOLOGY Tibial spines: male 2221, weak. Tml: male 0.38-0.40. Male palp: Figs. 3, 4, 9, 10, 13; the tibia bears a very stout spine. Diagnosis.— The male of D. stridulans is diagnosed by the form of the carapace lobe (Fig. 12), which has no holes or sulci; this character groups this species withD. procerus and D. kesimbus. D. stridulans is at once distinguished from D. kesimbus by the much more complicated embolus (Fig. 13 cf. Fig. 17), and by the form of the palpal tibia (Figs. 9, 10 cf. Figs. 16, 20). D. stridulans and D. procerus have closely similar palpal organs, but in D. stridulans the palpal patella is shorter (Fig. 9 cf. Fig. 15); the carapace lobes are also differently shaped, with the lobe in D. stridulans projecting well over the clypeus (Fig. 12 cf. Fig. 14). Distribution.— I have seen specimens from Utah only (Map IB). The females reported by Chamberlin and Ivie (1945) are Scotinotylus sanctus (Crosby). It has not been possible to locate the males reportedly taken in California (Chamberlin and Ivie 1945); it must be Figs. 9-15. -9, D. stridulans, male palp, ectal; 10, D. stridulans, male palpal tibia, dorsal; 11, D. stridulans, right male chelicera, lateral; 12, D. stridulans, male carapace, lateral; 13, D. stridulans, male palp, mesal; 14, D. procerus, male carapace, lateral; 15, D. procerus, male palp, ectal. Abbreviations: E, embolus; M, suprategular apophysis, membraneous part; T, tegulum (Scale lines 0.1 mm) MILLIDGE-GENUS DISEMBOLUS 267 regarded as uncertain that these males are D. stridulans , and the records are not included on the map. Natural History.— The male has been taken in September and October; nothing was recorded on habitat. Disembolus procerus, new species Figures 14, 15, 91, 92; Map IB Type.— Male holotype from Tieton River, 10 miles east of Rimrock, Washington, September 13, 1965 (J. and W. Ivie); deposited in AMNH. Description.— The male and female were taken together. Total length: female 1.45-1.55 mm, male 1.55 mm. Carapace: length: female/male 0.60-0.65 mm. Orange- brown to brown, with dusky markings and margins. The male carapace is raised anteriorly into a lobe (Fig. 14), which has no holes or sulci. Abdomen: grey to black; epigastric plates smooth. Sternum: brown, heavily suffused with black. Legs: brown to pale orange- brown. Tibial spines: female/male 2221, but rather weak in male. Tml: female/male 0.40-0.42. Male palp: Fig. 15; very similar to that of D. stridulans. Epigynum: Fig. 91. Internal genitalia: Fig. 92. Diagnosis.— The male is very similar to D. stridulans , and its diagnosis is dealt with under that species. The female of D. procerus is diagnosed by the epigynum (Fig. 91); the Figs. 16-21. -Male palps. 16, D. kesimbus, ectal; 17, D. kesimbus, mesal; 18, D. torquatus , ectal; 19, D. torquatus, mesal; 20, D. kesimbus, tibia, dorsal; 21, D. torquatus, tibia, dorsal. Abbreviations: AM, suprategular apophysis, auxiliary membrane; M, suprategular apophysis, membraneous part;TK, suprategular apophysis, tusk -like part. (Scale lines 0.1 mm) 268 THE JOURNAL OF ARACHNOLOGY small, rather widely spaced spermathecae separate it from most other species except D. galeatus (Fig. 90) and D. solanus (Fig. 89): see D. galeatus and D. solanus diagnoses. Distribution.— Known only from the type locality, Washington (Map IB). Natural History.— Both sexes were taken in September; nothing was recorded on habitat. Disembolus galeatus, new species Figures 90, 93; Map IB Type .—Female holotype from Lasa Falls, Little Cottonwood Canyon, Wasatch Mts., Salt Lake Co., Utah, April 19, 1961 (H. Levi); deposited in MCZ. Figs. 22-30. -Male palps. 22, D. alpha, ectal; 23, D. alpha, mesal; 24, D. beta, ectal; 25, D. beta, mesal; 26, D. sinuosus, ectal; 27, D. sinuosus, mesal; 28, D. alpha, tibia, dorsal; 29, D. beta, tibia, dorsal; 30, D. sinuosus, tibia, dorsal. Abbreviations: E, embolus; M, suprategular apophysis, mem- braneous part. (Scale lines 0.1 mm) MILLIDGE-GENUS DISEMBOLUS 269 Description.— Only the female is known. Total length: female 1.65 mm. Carapace: length: female 0.65 mm. Brown, with dusky markings and margins. Abdomen: black, epigastric plates smooth. Sternum: brown, suffused with black. Legs: brown, with the joints somewhat darker. Tibial spines: female 2221. Tml: female 0.43-0.47. Epigynum: Fig. 90; the outline of the posterior plate is helmet-shaped. Internal genitalia: Fig. 93. The similarity of the internal genitalia to those of D. procerus (Fig. 92), coupled with the location of capture, make it likely that this female may prove to be the unknown female of D. stridulans. Until the two sexes are captured together, however, this cannot be regarded as certain. Diagnosis.— The female is diagnosed by the epigynum (Fig. 90), which has widely separated spermathecae and a plate which is helmet shaped (though the outline may not always be very clear). More specimens are required to establish whether the difference from D. procerus epigynum (Fig. 91) is as clear as shown in the figures. A better know- ledge of the geographical distribution of D. galeatus and D. procerus may prove to be valuable for deciding the identity of doubtful females. Figs. 3 1-38. -Male palps. 31, D. sacerdotalis, ectal; 32, D. anguineus, ectal; 33, D. sacerdotalis, mesal; 34, D. anguineus, me sal; 35, D. anguineus, tibia, dorsal; 36, D. sacerdotalis, tibia, dorsal; 37, D. anguineus, tip of palp, dorsal view; 38, D. sacerdotalis, tip of palp, dorsal view. Abbreviations: C, cymbium; E, embolus; M, suprategular apophysis, membraneous part; TK, suprategular apophysis, tusk -like part. (Scale lines 0.1 mm) 270 THE JOURNAL OF ARACHNOLOGY Distribution.— Known only from the type locality, Utah (Map IB). Natural History.— The female was taken in April, at 2000 m. altitude, in douglas fir and cottonwood. Disembolus solanus , new species Figures 89, 107; Map IB Type. —Female holotype from Mix Canyon, Solano Co., California, March 12, 1960 (Parker and Menke); deposited in AMNH. Description.— Only the female is known. Total length: female 1.45 mm. Carapace: length: female 0.60 mm. Orange-brown. Abdomen: black; epigastric plates smooth. Sternum: yellow-brown. Legs: orange-brown. Tibial spines: female 2221. Tml: female 0.35. Epigynum: Fig. 89; the posterior plate is whitish in color. Internal genitalia: Fig. 107. Diagnosis.— The female is diagnosed by the epigynum (Fig. 89), the posterior plate of which has a shape rather different from that of the other species, and is whitish in color. More specimens are required to establish the extent of variation shown by the epigynum. Distribution.— This species is known from two localities in California (Map IB). Natural History.— The females were taken in February and March; nothing was recorded on habitat. Disembolus kesimbus (Chamberlin), new combination Figures 16, 17, 20, 55, 72, 94; Map ID Tapinocyba kesimba Chamberlin 1948:552 ( kesimba is assumed to be an adjective) Type.— The types were taken in Kaibab Forest (V. T. Ranch), Arizona, September 4, 1931 (R. V. Chamberlin). No holotype seems to have been designated, but a vial contain- ing male and female “types” is present in AMNH; a male has been selected and labelled as “lectotype.” Description.— Total length: female/male 1 .35-1.40 mm. Carapace: length: female /male 0.60 mm. Brown to orange-brown, with dusky markings. The male carapace is raised anteriorly, and the lobe has no holes or sulci (Fig. 55). Abdomen: grey; the epigastric plates have weak, closely spaced striae in the female, and clear, moderately spaced striae in the male. Sternum: brown, suffused with grey. Legs: orange-brown. Tibial spines: female 2221, male 0021. Tml: female 0.37-0.40, male 0.35. Male palp: Figs. 16, 17,20; the embolic division is less typical than in the other Disembolus species. Epigynum: Fig. 72; the plate is notched anteriorly: the small markings posterior to the notch are always present but variable in shape. Internal genitalia: Fig. 94. Diagnosis.— The male of D. kesimbus is diagnosed by the form of the carapace lobe (Fig. 55), which has no holes or sulci, and by the form of the palps (see D. stridulans diagnosis). The female of D. kesimbus is diagnosed by the epigynum (Fig. 72), which is easily recognized by the deep notch on the anterior margin of the plate, and by the two small curved markings posterior to the notch: these markings, though somewhat variable in shape, are always present. Distribution.— This species is known from several localities in Utah and one in Arizona (Map ID). MILLIDGE-GENUS DISEMBOLUS 211 Natural History.— Both sexes have been taken in September and October; nothing was recorded on habitat. Disembolus torquatus, new species Figures 1,5, 6, 18, 19,21,57,73,95; Map IB Type. -Male holotype from north-east of Fruitland, Idaho, November 24, 1940; deposited in AMNH. Description.— The male and female were taken together. Total length: female 1.45-1.60 mm, male 1.25-1.55 mm. Carapace: length: female/male 0.60-0.65 mm. Brown Figs. 39-48. -Male palps. 39, D. implicates, ectal; 40, D. implicates, mesal; 41, D. implexus, ectal; 42, D. implicates, tip of palp, dorsal view; 43, D. convolutus, tip of palp, dorsal view; 44, D. convolutus, ectal; 45, D. convolutus, mesal; 46, D. implicates, tibia, dorsal; 47, D. convolutus, tibia, ectal; 48, D. convolutus, tibia, dorsal. Abbreviations: C, cymbium; E, embolus; M, suprategular apophysis, membraneous part; TK, suprategular apophysis, tusk -like part. (Scale lines 0.1 mm) 272 THE JOURNAL OF ARACHNOLOGY to orange-brown with blackish markings and margins. The male carapace has a large lobe which has well marked sulci and holes (Fig. 57). Abdomen: grey to black; epigastric plates smooth. Sternum: brown, suffused with black, particularly on margins. Legs: brown to orange-brown. Tibial spines: female 2221, male spineless. Tml: female 0.38-0.42, male 0.36-0.40. Male palp: Figs. 1, 5, 6, 18, 19, 21. Epigynum: Fig. 73; the plate has a notch anteriorly. Internal genitalia: Fig. 95. Diagnosis.— The male of D. torquatus is diagnosed by the carapace lobe, which has large holes and sulci (Fig. 57), coupled with the form of the tibial apophysis and of the SA, which has a curved hook distally (Fig. 18). Confirmation is given by the palpal organs (Fig. 19). The female is diagnosed by the epigynum, which differs from other species in the presence of a broad notch on the anterior margin of the plate (Fig. 73). Distribution.— Known only from the type locality, Idaho (Map IB). Natural History.— Both sexes were taken in September; nothing was recorded on habitat. Disembolus alpha (Chamberlin), new combination Figures 22, 23, 28, 56, 83, 96; Map 1 A Tapinocyba alpha Chamberlin 1948:550 Type.— Male and female types from Dry Creek Canyon, Salt Lake City, Utah, October 22, 1932; in AMNH, examined. A holotype has not been designated. A male lectotype has been selected and deposited in AMNH. Description.— Total length: female 1.1 mm, male 1.05-1.2 mm. Carapace: length: fe- male 0.50 mm, male 0.50-0.54 mm. Yellow-brown, with dusky markings and margins. 53 54 52 Figs. 49-54. -Male palps. 49, D. phanus, ectal; 50, D. lacunatus, tibia, ectal; 51, D. phanus, tibia, dorsal; 52, D. phanus, mesal; 53, D. phanus, tip of palp, dorsal view; 54, D. lacunatus, tip of palp, dorsal view. Abbreviations: C, cymbium; E, embolus; M, suprategular apophysis, membraneous part. (Scale lines 0.1 mm) MILLIDGE-GENUS DISEMBOLUS 273 The male carapace has no lobe, but there is a tiny hole behind the lateral eyes (Fig. 56). Abdomen: grey, suffused with whitish yellow; epigastric plates smooth. Sternum: yellow, suffused with grey. Legs: pale yellow. Tibial spines missing on all specimens. Tml: female 0.40, male 0.38-0.40. Male palp: Figs. 22, 23, 28; the embolus is less convoluted than in many of the Disembolus species. Epigynum: Fig. 83. Internal genitalia: Fig. 96. Diagnosis.— The male of D. alpha is diagnosed at once by the absence of a lobe on the carapace, and the presence of a small hole behind the lateral eyes (Fig. 56). Confirmation is given by the form of the palpal tibia (Fig. 22), by the relatively simple form of the embolus (Fig. 23) and by the small size of the spider. The palpal tibia and ED are similar to those of D. beta (Fig. 24, 25) and D. sinuosus (Fig. 26, 27), but these two species are at once distinguished fromD. alpha by the presence of a carapace lobe (Figs. 58, 59). The female of D. alpha is diagnosed by the epigynum, which has two clear banana-shaped markings within the plate (Fig. 83), coupled with the small size of the spider. Distribution.— Known only from the type locality, Utah (Map 1 A). Natural History.— Both sexes were taken in October; nothing was recorded on habitat. Figs. 55-62. -Male carapaces. 55, D. kesimbus, lateral; 56, D. alpha, lateral; 57, D. torquatus, lateral; 58, D. sinuosus, lateral; 59, D. beta, lateral; 60, D. anguineus, lateral; 61, D. sacerdotalis , lateral (after Crosby and Bishop, 1933); 62, D. anguineus, dorsal. (Scale lines 0.2 mm) 274 THE JOURNAL OF ARACHNOLOGY Disembolus beta, new species Figures 24, 25, 29, 59; Map 1C Type.— Male holotype from Dry Canyon, Salt Lake City, Utah, October 22, 1932 (W. Ivie); deposited in AMNH. Description.— Only the male is known. Total length: male 1.1-1.15 mm. Carapace: length: male 0.50 mm. Pale orange-brown, with faint dusky markings and margins. The male carapace is raised into a lobe with holes and sulci (Fig. 59), and the clypeus projects. Abdomen: yellowish-grey; epigastric plates smooth. Sternum: yellow-brown with grey margins. Legs: yellow-brown. Tibial spines absent. Tml: male 0.40. Male palp: Figs. 24, 25, 29; the embolus is less convoluted than in many of the Disembolus species. Diagnosis.— The male of D. beta is diagnosed by the presence of a carapace lobe with holes and sulci (Fig. 59), coupled with the shape of the tibial apophysis (Fig. 24); these characters group it with D. sinuosus. D. beta is separated from D. sinuosus by its smaller size (ca. 1.1 mm cf. 1 .4 mm), by the somewhat shorter tibial apophysis (Fig. 29, cf. Fig. 30) and by the smaller size of the embolic coil (Fig. 25 cf. Fig. 27). D. beta has the palpal organs and tibia very similar to those of D. alpha , but differs from this species in the presence of the carapace lobe (Fig. 59 cf. Fig. 56). Distribution. -Known only from two localities in Utah (Map 1C); the holotype male appears to have been taken in the same locality and on the same date as D. alpha. 70 ' ' 71 Figs. 63-71. -Carapaces. 63, D. implicatus, male, lateral; 64, D implexus, male, lateral; 65, D. convolutus, male, lateral; 66, D. phanus, male, lateral; 67, D. lacunatus, male, lateral; 68, D. implexus, male, dorsal; 69, D. implicatus, female, lateral; 70, D. phanus, female, lateral; 71, D. vicinus, female, lateral. (Scale lines 0.2 mm) MILLIDGE-GENUS DISEMBOLUS 275 Natural History.— The males were taken in October; nothing was recorded on habitat. Disembolus sinuosus , new species Figures 26, 27, 30, 58, 78, 97; Map 1A Type.— Male holotype from the summit of Mt. Washburn, Wyoming, August 13, 1940 (W. Ivie); deposited in AMNH. Description.— The male and female were taken together. Total length: female/male 1.40 mm. Carapace: length: female/male 0.60 mm. Dark brown, with blackish markings and margins. The male carapace is raised into a lobe, with holes and sulci (Fig. 58), and the clypeus projects. Abdomen: black; epigastric plates smooth. Sternum: brown, heavily suffused with black. Legs: brown to orange-brown. Tibial spines: female 2221, male, Figs. 72-82.-Epigyna. 72, D. kesimbus, ventral; 73, D. torquatus, ventral; 74, D. convolutus, ventral; 75, D. anguineus, ventral; 76, D. amoenus, ventral; 77, D. amoenus, posterio-ventral; 78, D. sinuosus, ventral; 79, D. concinnus, ventral; 80, D. hyalinus, ventral; 81, 82, D. corneliae, ventral. (Scale lines 0.1 mm) 276 THE JOURNAL OF ARACHNOLOGY spines missing (or absent). Tml: female 0.40-0.44, male 0.47. Male palp: Figs. 26, 27, 30. Epigynum: Fig. 78. Internal genitalia: Fig. 97. Diagnosis.-/), sinuosus male is diagnosed by the presence of the carapace lobe with holes and sulci (Fig. 58), coupled with the form of the tibial apophysis (Fig. 26); these characters group it with D. beta, and its diagnosis is dealt with under that species. The female of D. sinuosus is diagnosed by the epigynum, which has a small dark colored wedge-shaped marking just anterior to the plate (Fig. 78), and clear markings across and within the plate. Distribution.— Known only from the type locality, Wyoming (Map 1 A). Natural History.— Both sexes were taken at ca. 3000 m in August; nothing was recorded on habitat. Disembolus sacerdotalis (Crosby and Bishop), new combination Figures 31 , 33, 36, 38, 61 ; Map 1 A Cochlembolus sacerdotalis Crosby and Bishop 1933:167 Type.— Male holotype from Karner, Albany Co., New York, March 24, 1923; in AMNH, examined. Description.— This species is known only from the holotype. This specimen is in bad condition, with all the legs missing and the carapace damaged; fortunately the palps are present. The following brief and incomplete description is therefore based for the most part on the data given by Crosby and Bishop (1933). Total length: male 1.90 mm. Carapace: length: male 0.90 mm. Yellow-brown. Carapace steeply rising to a small lobe anteriorly, with small holes and sulci (Fig. 61); the clypeus is strongly convex. Abdomen: dark grey; epigastric plates smooth. Sternum: dull yellow, with dusky markings. Legs: dusky orange-yellow. Male palp: Figs. 31, 33, 36, 38. Diagnosis.— The male is diagnosed by the form of the carapace, which has a lobe with holes and sulci (Fig. 61), and by the tibial apophysis which is relatively long and hooked distally (Fig. 31). Confirmation is given by the form of the ED (Fig. 33) and of the SA (Fig. 38). This species is relatively large (total length ca. 2 mm). Distribution.— Known only from the type locality, New York (Map 1 A). Natural History.— The male was taken in March, by sifting leaf mold. Disembolus anguineus, new species Figures 2, 7, 8, 32, 34, 35, 37, 60, 62, 75, 101; Map 1A Type.— Male holotype from Chiricahua Mts., Cochise Co., Arizona, December 17, 1954 (K. W. Haller); deposited in AMNH. Description.— The male and female were taken together. Total length: female 1.65 mm, male 1.45-1.55 mm. Carapace: length: female 0.65-0.75 mm, male 0.65-0.70 mm. Orange-brown, with dusky markings and margins. The male carapace is raised into a lobe (Figs. 60, 62), with holes and sulci. Abdomen: yellowish grey; epigastric plates smooth. Sternum: orange, heavily suffused with black. Legs: orange-brown. Tibial spines: female 2221, male, spines missing (or absent). Tml: female 0.40-0.45, male 0.45-0.50. Male palp: Figs. 2, 7, 8, 32, 34, 35, 37. Epigynum: Fig. 75. Internal genitalia: Fig. 101. Diagnosis.— The male of D. anguineus is diagnosed by the form of the carapace, which has a lobe with holes and sulci, and a fairly strongly projecting clypeus (Figs. 60, 62), MILLIDGE-GENUS DISEMBOLUS 277 coupled with the form of the tibial apophysis (Fig. 32). Confirmation is afforded by the form of the ED (Fig. 34) and of the SA (Fig. 37). The female of D. anguineus is diagnosed by the epigynum, which has a small dark-colored knob on the anterior margin, with the spermathecae close together (Fig. 75); this epigynum is unlikely to be confused with that of any other Disembolus species. Distribution.— This species is known from Utah, Arizona and New Mexico (Map 1A). Natural History.— Both sexes have been taken in October, December and March; it seems probable that the chief period of maturity is in autumn and winter. In New Mexico it was taken in March in pitfall traps (at Los Alamos), and also at ca. 2500 m in October. Disembolus implicatus , new species Figures 39, 40, 42, 46, 63, 69, 84, 85, 98; Map 1C Type.— Male holotype from Cobble Rest, Upper Provo River, Utah, July 30, 1936 (W. Ivie); deposited in AMNH. Description.— The male and female were taken together. Total length: female 1.45-1.55 mm, male 1.40 mm. Carapace: length: female/male 0.60-0.65 mm. Brown, with dusky markings and margins. The female carapace is well elevated behind the eyes (Fig. 69). The male carapace bears a lobe which has prominent holes and sulci behind the lateral eyes (Fig. 63). Abdomen: grey to black; epigastric plates smooth or with very weak striae. Sternum: brown, suffused with black. Legs: brown to orange-brown. Tibial Figs. 83-91.— Epigyna, ventral. 83, D. alpha ; 84, 85, D. implicatus ; 86, D. phanus', 87, D. lacteus', 88, D. vicinus ; 89, D. solanus’, 90, D. galeatus’, 91, D. procerus. (Scale lines 0.1 mm) 278 THE JOURNAL OF ARACHNOLOGY spines: female 2221, male 001 1. Tml: female 0.38-0.50, male 0.40-0.44. Male palp: Figs. 39, 40, 42, 46. Epigynum: Figs. 84, 85; two females taken in Colorado show small differences in the epigynum (Fig. 85), but in the absence of males are assumed to be D. implicatus. Internal genitalia: Fig. 98. Diagnosis.— The male of D. implicatus is diagnosed by the presence of the carapace lobe with large holes and sulci (Fig. 63), coupled with the form of the tibial apophysis, which has a forward-directed point distally (Fig. 39). These characters group it with D. implexus and D. convolutus. D. implicatus is very similar to D. implexus, and is distin- guished only by the form of the carapace lobe (Fig. 63 cf. Figs. 64, 68) and by very small differences in the SA. D. implicatus and D. implexus are readily distinguished from D. convolutus by the form of the ED, which in the latter species is more elongated and has the embolic coil smaller in diameter (Fig. 40 cf. Fig. 45), and by the form of the palpal tibia (Fig. 46 cf. Figs. 48); the carapace lobe of D. convolutus is fairly similar to that of D. implexus (Figs. 64, 65). The female of D. implicatus is diagnosed by the epigynum (Figs. 84, 85), which is one of the simplest of the genus. This epigynum is very similar to that of D. phanus (Fig. 86), and could easily be confused with it. These two species can be distinguished, however, by the more abrupt elevation of the carapace immediately behind the eyes in D. phanus (Fig. 70 cf. Fig. 69). Distribution.— Known from two localities in Utah and from Colorado (Map 1C). Natural History.— Males were taken in July, females in July, August and October; the chief maturity period is probably in summer. Nothing was recorded on habitat. 95 100 Figs. 92-100.- Internal genitalia, females, ventral. 92, D. procerus ; 93, D. galeatus ; 94, D. kesimbus ; 95, D. torquatus ; 96, D. alpha ; 97, D. sinuosus ; 98, D. implicatus ; 99, D. phanus ; 100,/). vicinus. (Scale lines 0.1 mm) MILLIDGE-GENUS DISEMBOLUS 279 Disembolus implexus, new species Figures 41, 64, 68; MaplC Type. -Male holotype from Fish Lake, Utah, September 4, 1929 (Chamberlin and Gertsch); deposited in AMNH. Description.— Only the male is known. Total length: male 1.55 mm. Carapace: length: male 0.70 mm. Orange-brown, with lobe pale yellow; raised into large lobe, which bears long hairs anteriorly (Figs. 64, 68). Abdomen: grey-black; epigastric plates smooth. Sternum: yellow, suffused with black. Legs: yellow. Tibial spines: male 1121, very short and weak. Tml: male 0.45. Male palp: Fig. 41; almost identical with that of D. implicatus. Diagnosis.— The male of D. implexus is very similar to D. implicatus , and its diagnosis is dealt with under that species. Distribution.— Known only from the type locality, Utah (Map 1C). Natural History.— The male was taken in September; nothing was recorded on habitat. Disembolus convolutus, new species Figures 43, 44, 45, 47, 48, 65, 74, 102; Map 1C Type.— Male holotype from 2 miles east of Rufus, Oregon. November 25, 1940 (W. Ivie); deposited in AMNH. Description.— The male and female were taken together. Total length: female 1.45-1.55 mm, male 1.25-1.35 mm. Carapace: length: female/male 0.60 mm. Orange- brown with dusky markings and margins. The male carapace has a lobe with holes and sulci (Fig. 65). Abdomen: grey to yellow-grey; epigastric plates smooth. Sternum: orange- brown, suffused with black. Legs: orange-brown. Tibial spines: female 2221, male 0011. Tml: female 0.53-0.56, male 0.50-0.55. Male palp: Figs. 43, 44, 45, 47, 48. Epigynum: Fig. 74. Internal genitalia: Fig. 102. Diagnosis. -The male of D. convolutus is diagnosed by the presence of the carapace lobe with holes and sulci (Fig. 65), coupled with the form of the tibial apophysis (Fig. 47). These characters group it with D. implicatus and D. implexus , and its diagnosis is dealt with under D. implicatus. The female of D. convolutus is diagnosed by the epigynum, which is simple apart from the presence of a V-shaped ridge on the anterior margin (Fig. 74); this epigynum is unlikely to be confused with that of any other known species of the genus. Distribution.— Known only from the type locality, Oregon (Map 1C). Natural History.— Both sexes were taken in November; nothing was recorded on habitat. Disembolus phanus (Chamberlin), new combination Figures 49, 51,52,53,66,70,86,99; Map ID Tapinocyba (?) phana Chamberlin 1948:553. It is assumed that phana is an adjective. Type.— Female holotype from 4 miles N.E. of McCall, Idaho, October 18, 1944 (W. Ivie); in AMNH, examined. Description.— The male, which was taken with females, is described for the first time. Total length: female 1.45-1.65 mm, male 1.45-1.55 mm. Carapace: length: female/male 280 THE JOURNAL OF ARACHNOLOGY 0.60-0.65 mm. Orange-brown, with blackish markings and margins. The female carapace is abruptly elevated behind the eyes (Fig. 70). The male carapace is raised into a large lobe, which projects forwards over the ocular area (Fig. 66); there are sulci and deep wide holes behind the lateral eyes. Abdomen: grey to black; epigastric plates smooth. Sternum: almost black. Legs: brown to orange-brown. Tibial spines: female 2221, male 0021. Tml: female 0.37-0 45, male 0.40-0.45. Male palp: Figs. 49, 51, 52, 53; the tibia bears one very stout spines. Epigynum: Fig. 86. Internal genitalia: Fig. 99. Diagnosis.—/), phanus male is diagnosed by the presence of a large lobe with holes and sulci (Fig. 66), coupled with the form of the palpal tibia (Fig. 49), which has a forward- directed point distally and bears a very stout spine; these characters place D. phanus with its close relative D. lacunatus. These two species have almost identical palpal organs, but show small differences in the SA (Fig. 53 cf. Fig. 54), and there are also small differences in the palpal tibia (Fig. 49 cf. Fig. 50); the carapace lobes are on the contrary quite different in form (Fig. 66 cf. Fig. 67). The female of D. phanus is diagnosed by the epigynum (Fig. 86), which is very similar to that of D. implicatus; these two species are distinguished by the more abrupt post-ocular elevation of the carapace in D. phanus (Fig. 70 cf. Fig. 69). Distribution.— This species is known from Idaho and Montana (Map ID). Natural History. -Both sexes were taken in October; nothing was recorded on habitat. Disembolus lacunatus , new species Figures 50, 54, 67; Map ID Type.— Male holotype from Big Wood River, 19 miles N. of Ketchum, Idaho. August 25, 1941 (Chamberlin and Piemeisel); deposited in AMNH. Description.— Only the male is known. Total length: male 1.5 mm. Carapace: length: male 0.70 mm. Orange-brown; raised anteriorly into a lobe, with sulci and large holes (Fig. 67); the clypeus does not project. Abdomen: yellowish grey; epigastric plates smooth. Sternum: yellow, suffused with black. Legs: yellow. Tibial spines missing (or absent). Tml: male 0.52. Male palp: Figs. 50, 54; the tibia bears a very stout spine. The palpal organs are practically identical with those of D. phanus , except for a small dif- ference in the tip of the membraneous part of the SA (Fig. 54). Diagnosis.—/), lacunatus male is closely related to D. phanus, and is distinguished by the form of the carapace lobe (see D. phanus diagnosis). Distribution.— Known only from the type locality, Idaho (Map ID). Natural History.— The male was taken in August; nothing was recorded on habitat. Disembolus lacteus, new species Figures 87, 103; Map 1C Type. -Female holotype from Fallen Leaf Lake, California, August 24, 1953 (J. D. Lattin); deposited in AMNH. Description.— Only the female is known. Total length: female 1.55 mm. Carapace: length: female 0.75 mm. Orange-brown with dusky markings and black margins. Abdomen: black; epigastric plates smooth. Sternum: brownish black. Legs: orange- brown. Tibial spines: female 2221. Tml: female 0.50. Epigynum: Fig. 87; the posterior plate is milky white in color, and there is a yoke-like marking across the epigynum MILLIDGE-GENUS DISEMBOLUS 281 anterior to the plate. The spermathecae are rather widely separated. Internal genitalia: Fig. 103. Diagnosis. -7). lacteus female is diagnosed by the epigynum (Fig. 87), the posterior plate of which is more or less devoid of markings and is milky white in color. Distribution.-Known only from the type locality, California (Map 1C). Natural History. -The female was taken in August; nothing was recorded on habitat. Disembolus vicinus , new species Figures 71, 88, 100; Map 1C Type.— Female holotype from Grantsville, Tooele Co., Utah, April 24, 1961 (H. Levi); deposited in MCZ. Description. -Only the female is known. Total length: female 1.65 mm. Carapace: length: female 0.60 mm. Orange-brown, with dusky markings and margins. The female carapace is moderately raised behind the eyes (Fig. 71); this degree of elevation in the female indicates that the male carapace will probably have a fairly large lobe. Abdomen: black; epigastric plates smooth. Sternum: orange, heavily suffused with black. Legs: orange-brown. Tibial spines: female 2221. Tml: female 0.40-0.42. Epigynum: Fig. 88. Internal genitalia: Fig. 100. Fig. 101-108. -Internal genitalia, females, ventral. 101, D. anguineus ; 102, D. convolutus; 103, D. lacteus ; 104, D. amoenus; 105, D. concinnus\ 106, D. corneliae; 107,7). solanus', 108,7). hyalinus (Scale lines 0.1 mm) 282 THE JOURNAL OF ARACHNOLOGY Diagnosis. -D. vicinus female is diagnosed by the epigynum (Fig. 88), which has the posterior plate more or less devoid of markings and the spermathecae close together. Distribution.— Known only from the type locality, Utah (Map 1C). Natural History. -The female was taken in April, in sage trash at 1300 m altitude. Disembolus amoenus, new species Figures 76, 77, 104; Map ID Type.— Female holotype from Cumberland Pass, 12,500 ft., Sawatch Range, Gunnison Co., Colorado, July 13, 1957 (H. and L. Levi); deposited in MCZ. Description.— Only the female is known. Total length: female 1.35-1.70 mm. Carapace: length: female 0.60-0.65 mm. Orange-brown to deep brown, with dusky mark- ings and margins. Abdomen: black; epigastric plates smooth but dark in color. Sternum: brown, suffused with black. Legs: orange-brown to brown. Tibial spines: female 2221. Tml: female 0.40-0.44. Epigynum: Figs. 76, 77; when viewed from behind (Fig. 77), the small ledge is invisible. Internal genitalia: Fig. 104. Diagnosis.— D. amoenus female is diagnosed by the epigynum, which has a small and rather indistinct V-shaped marking just anterior to the plate (Fig. 76); when viewed somewhat from behind, however, the marking is not visible (Fig. 77). In the Key, D. sinuosus falls into the same group with D. amoenus , but the epigyna of these two species are sufficiently different in appearance (Fig. 76 cf. Fig. 78) to make confusion unlikely. Distribution.— Known from two localities in Colorado (Map ID). Natural History.— The female has been taken in July and September; one habitat quoted is under stones and rocks, at 3800 m altitude (Cumberland Pass), where it was taken in company with Scotinotylus majesticus (Chamberlin and Ivie). Disembolus concinnus, new species Figures 79, 105; Map ID Type.— Female holotype from Southwestern Research Station, Arizona, September 25, 1956 (A. M. Nadler); deposited in AMNH. Description.— The two females taken were accompanied by a sub-adult male almost at the final moult: the palpal organs and the tibial apophysis were visible through the integument. The data given for the male are based on this sub-adult specimen. Total length: female 1.70-1.80 mm, male 1.60 mm. Carapace: length: female 0.75-0.80 mm, male 0.65 mm. Brown, with dusky markings and margins. The sub-adult male has no lobe or sulci visible on the carapace, and it is probable therefore that the adult will have at most a shallow lobe. Abdomen: black; epigastric plates smooth. Sternum: yellow-brown, heavily suffused with black. Legs: yellow-brown. Tibial spines: female/male 2221. Tml: female 0.46-0.50, male 0.42. Male palp: the palpal organs, seen through the integument, are somewhat obscure, but seem to be similar to those of D. implicatus\ the palpal tibia also resembles that of D. implicatus. Epigynum: Fig. 79; when viewed from behind, the mantle is scarcely visible, and the epigynum has a similar appearance to Fig. 77. Internal genitalia: Fig. 105. Diagnosis. -D. concinnus female is diagnosed by the epigynum, which has a large, dark-colored, V-shaped mantle anterior to the plate (Fig. 79); viewed more from behind, the epigynum approaches in appearance that of D. amoenus (Fig. 77). There are clear MILLIDGE-GENUS DISEMBOLUS 283 markings within the plate, rather similar to those of D. alpha ; since there are only two females known, it is uncertain to what extent these markings vary in shape. D. concinnus is one of the larger species of the genus (total length ca. 1.8 mm). Distribution.— Known only from the type locality, Arizona (Map ID). Natural History.— The females were adult at the end of September; the male was then ready for its final moult, and would probably have become adult in October. Nothing was recorded on habitat. Disembolus corneliae (Chamberlin and Ivie), new combination Figures 81, 82, 106; Map 1A Soudinus corneliae Chamberlin and Ivie 1944:77 Type.— Female holotype from Demorest, Georgia, April 26, 1943 (W. Ivie); in AMNH, examined. Description.— Only the female is known. Total length: female 1.95-2.0 mm. Carapace: length: female 0.75-0.80 mm. Orange-brown to brown, with dusky markings and black margins. Abdomen: grey to black; epigastric plates smooth. Sternum: brown, suffused with black. Legs: brown to orange-brown. Tibial spines: female 2221. Tml: female 0.50-0.55. Epigynum: Figs. 81, 82. The posterior plate is somewhat variable in appearance, depending probably on the transparency of the integument, and the size of the mantle also shows small variations; when viewed from behind, the mantle is much less visible. Internal genitalia: Fig. 106. The material described as D. corneliae may prove to be the female of D. sacerdotalis; these two species have in common an unusually large size (for the genus), and both are found in the eastern part of the continent. Capture of the two sexes together, or at least in the same locality, would be necessary to confirm this suggestion. Diagnosis.-/), corneliae female is diagnosed by the epigynum, which has a large, dark-colored mantle, trapezoidal in shape, anterior to the plate (Figs. 81, 82). There are clear markings within the plate, and although these show some variation they seem to be characteristic of the species. D. corneliae is one of the larger species of the genus (total length ca. 2 mm). Distribution.-This species is known from New Jersey, Indiana, Georgia and S. Carolina (Map 1 A). Natural History.— The female has been taken in April and in “summer.” Nothing was recorded on habitat. Disembolus hyalinus, new species Figures 80, 108; Map 1A Type.— Female holotype from Lake Agnes, Alberta, August 4, 1927; deposited in AMNH. This specimen, in the S. C. Bishop Collection, consists of an abdomen only. Description. -Although only the abdomen is present, I have little hesitation in record- ing this as a new species, since the epigynum indicates fairly conclusively that we are dealing with an undescribed species of Disembolus. The description is necessarily incom- plete. Total length: probably ca. 2 mm, based on the abdomen length. Abdomen: length: 1.3 mm; grey-black, epigastric plates smooth. Epigynum: Fig. 80. Internal genitalia: Fig. 108. 284 THE JOURNAL OF ARACHNOLOGY Diagnosis.—/), hyalinus is diagnosed by the epigynum (Fig. 80). There is a dark line anterior to the plate, but no ledge or mantle; the plate is very glassy, and the markings visible are different from those of any other known species. D. hyalinus is probably one of the larger species of the genus. Distribution. -Known only from the type locality, Alberta (Map 1 A). Natural History.— The female was taken in August; nothing was recorded on habitat. ACKNOWLEDGEMENTS I am indebted to the following colleagues for the loan of material: Dr. N. I. Platnick, American Museum of Natural History, New York; Prof. H. W. Levi, Museum of Compara- tive Zoology, Harvard University; Dr. C. D. Dondale, Agriculture Canada, Biosystematics Research Institute, Ottawa. Drs. Platnick and Dondale gave some valuable help with the mapping, and Mr. F. R. Wanless and Mr. P. Hillyard (British Museum, Natural History) and Mr. G. H. Locket (Stockbridge) provided me with some of the literature. LITERATURE CITED Chamberlin, R. V. 1948. On some American spiders of the family Erigonidae. Ann. Ent. Soc. America, 41:483-562 Chamberlin, R. V. and W. Ivie. 1933. Spiders of the Raft River Mountains of Utah. Bull. Univ. Utah (Biol.), 23(4): 1-53 Chamberlin, R. V. and W. Ivie. 1944. Spiders of the Georgia region of North America. Bull. Univ. Utah (Biol.), 35(9): 1-267 Chamberlin, R. V. and W. Ivie. 1945. Erigonid spiders of the genera Spirembolus, Disembolus and Bactroceps. Trans. Connecticut Acad. Arts Sci., 36:215-235 Crosby, C. R. and S. C. Bishop. 1933. American spiders: Erigoneae, males with cephalic pits. Ann. Ent. Soc. America, 26(1): 105-182 Millidge, A. F. 1980. The erigonine spiders of North America. Part 2. The genus Spirembolus Chamberlin (Araneae: Linyphiidae). J. Arachnol., 8(2): 109-158. Millidge, A. F. 1981. The erigonine spiders of North America. Part 3. The genus Scotinotylus Simon (Araneae: Linyphiidae). J. Arachnol., 9: 167-213 Manuscript received February 1980, revised May 1980. Coyle, F. A. 1981. Effects of clearcutting on the spider community of a Southern Appalachian forest. J. Arachnol., 9:285-298. EFFECTS OF CLEARCUTTING ON THE SPIDER COMMUNITY OF A SOUTHERN APPALACHIAN FOREST Frederick A. Coyle Department of Biology Western Carolina University Cullowhee, North Carolina 28723 ABSTRACT In order to examine the effects of clearcutting on the spider community of a southern Appalachian forest, spider populations in a mature forest, an adjacent clearcut site, and two other clearcut sites were sampled during the summer of 1976 with four techniques: pitfall trapping, Tullgren funnel extraction of litter spiders, sweep netting, and hand collecting. Clearcutting resulted in a marked reduction in spider abundance and a small decrease in the number of spider species. Spider species diversity increased, owing to a marked increase in the evenness component of diversity. While clearcut- ting greatly reduced the abundance and number of species of web building spiders (both ground- dwelling and aerial species), there was apparently little or no reduction in numbers of hunting spiders. In addition, the number of hunting spider species increased. It is postulated that microclimate changes resulting from removal of the forest canopy and reduced litter thickness were primarily responsible for the decline in abundance and diversity of web builders. INTRODUCTION Growing interest in managing forests as ecosystems which are valued for many func- tions has increased the need for understanding the impact of forest management practices on a wide variety of organisms (Boyce 1977), including spiders. Mounting evidence indi- cates that the population density, behavior, and population dynamics of spiders are such that these predators are collectively an important stabilizing agent of terrestrial arthropod populations (Breymeyer 1966, Moulder and Reichle 1972, Turnbull 1973, Riechert 1974, Enders 1975) and thus may be an important factor in total ecosystem stability. The current study was initiated to examine how the population size, species diversity, and guild composition of the spider component of a southern Appalachian forest community are affected by clearcutting. In addition to helping understand the adaptations and response capabilities of different types of spiders, it may contribute to more informed forest management. STUDY SITES Four study sites were chosen, all in the Highlands Ranger District of the Nantahala National Forest in the mountains of southwestern North Carolina. The sites (Table 1) 286 THE JOURNAL OF ARACHNOLOGY Table 1. -General characteristics of the study sites. Age of clearcut in summers since clearcutting Size in ha Slope Dominant aspect Mean elevation in m Ellicott Rock forest 8 5°-20° 100° 860 Ellicott Rock clearcut 2nd 8 5°-12° 100° 880 Buck Creek clearcut 1st 10 20° -30° 20° 975 Horse Cove clearcut 5 th 16 2°-10° 130° 950 include one area of mature forest, an adjacent, topographically very similar area in its second summer following clearcutting, and two separate areas representing the first and fifth summers following clearcutting. All clearcutting was performed during the winter. Ellicott Rock forest site.— The forest site, located near Ellicott Rock, is occupied by a mature pine-hardwood community bissected by a narrow, weakly developed cove forest community along a small stream. The dominant tree species (in order of decreasing importance values, which are based upon relative frequency and relative dominance values obtained by the point quarter method) are white pine ( Pinus strobus), white oak ( Quer - cus alba), sourwood ( Oxydendrum arboreum), black oak ( Quercus velutina), red maple ( Acer rubrum), and scarlet oak (Quercus coccinea) in the pine-hardwood community; and red maple, Canadian hemlock ( Tsuga canadensis), sourwood, black gum (Nyssa sylvatica ), white pine, tulip tree (Liriodendron tulipifera), white oak, and holly (Ilex opaca) in the cove forest community. The moderately dense understory of the pine-hardwood com- munity contained dogwood (Cornus florida), numerous young of canopy species, and shrubs such as huckleberry (Gaylussacia ursina), scattered mountain laurel (Kalmia lati- folia), and scattered Rhododendron maximum. Dense stands of R. maximum dominated the cove forest understory. Leaf litter depth in the forest ranged from 1 to 15 cm (n = 10), with a mean depth of 6.5 cm. Clearcut sites.— Vegetation at the clearcut study sites was analyzed by nested quadrat sampling (Horn 1976). The Buck Creek site was sampled the summer of the spider study, but the Ellicott Rock and Horse Cove clearcuts were sampled the previous summer. In order of decreasing importance value, the five most important species of woody plants over 0.5 m tall at each clearcut site were: Ellicott Rock — huckleberry, red maple, sourwood, greenbriar (Smilax rotundi folia), and pignut hickory (Cary a ovalis)\ Buck Creek — American chestnut (Castanea dentata), huckleberry, dogwood, greenbriar, and buffalo nut (Pyrularia pubera); Horse Cove — blackberry (Rubus allegheniensis), huckle- berry, tulup tree, mountain laurel, and spicebush (Calycanthus floridus). Subjective observations indicate that the foliage density and percentage of shaded ground increased significantly with the age of the clearcut sites, with the Buck Creek site having the lowest and the Horse Cove site having the greatest foliage density. Litter depth varied as follows: Ellicott Rock, range = 1-9 cm (n = 10), mean = 4.2 cm; Buck Creek, range = 2-10 cm (n = 10), mean = 4.8 cm; Horse Cove, range = 1-8 cm (n = 10), mean = 3.9 cm. Comparison of both the pre-logging importance values (calculated from timber cruise curveys [Horn 1976]) of tree species at the clearcut sites and the importance values of potential canopy species in the clearcut site quadrats (Horn 1976) with the Ellicott Rock forest site vegetation analysis indicates that this Ellicott Rock forest is botanically the same as the pre-clearcut forest at the adjacent clearcut site, but that the pre-clearcut COYLE-EFFECTS OF FOREST CLEARCUTTING ON SPIDERS 287 forests at the Buck Creek and Horse Cove sites are markedly different from the Ellicott Rock forest and from one another. METHODS Four different collecting techniques were used to sample the spider populations at these study sites during the summer of 1976. An attempt was made to distribute the samples evenly over each site and to sample from each type of microhabitat at each site. No samples were collected within 20 m of the edge of any study site. Eight 73 mm diameter sheltered pitfall traps containing an ethylene glycol-detergent mixture were set on each study site on June 25 and were then emptied and reset at three week intervals during the following 15 weeks. Ten 0.25 m2 samples of leaf litter (down to the mineral soil) were collected from each study site at fairly regular intervals between June 16 and August 20 and were processed in large Tullgren funnels. Eight daytime sweep net samples of 50 sweeps apiece were obtained from vegetation between 0.2 m and 2.0 m above ground level at each site between June 29 and July 2. Four hours of intensive daytime hand collecting was performed at each site between June 22 and July 9, with an addi- tional 30 minutes of intensive daytime hand collecting at each site on October 2. Search time was divided equally between the ground stratum and the aerial stratum (branches and leaves above the ground). Ground and aerial spiders were placed in separate collecting vials. RESULTS Table 2 shows the number of individuals of each spider taxon collected at each site. A total of 1729 individuals representing at least 134 species and 23 families were collected from all sites. Caution must be used in interpreting these data. The sampling was not strictly random. Temporal bias exists favoring species that are more abundant or active during the summer and the daytime. Each of the collecting techniques used collects certain kinds of spiders more effectively than others (Turnbull 1973). In addition, the forest canopy was not sampled. Because of these biases, the data cannot be expected to closely represent the real population densities or total number of species at any site. Nevertheless, since the collecting at all sites was concurrent and involved the same tech- niques and effort, the data should reflect with reasonable accuracy between-site differ- ences in the spider communities. It is very important to emphasize that, since the pre-clearcut plant communities at the Buck Creek and Horse Cove sites were different from the Ellicott Rock forest, these two clearcut sites cannot be treated as different time points of the succession one would expect to witness at the Ellicott Rock site. Consequently, the analysis of results will focus primarily on the spider data from the Ellicott Rock sites, and will be based on the assumption, supported by the vegetation analyses, that significant differences between the spider samples from these two sites are primarily the results of clearcutting. Effects of clearcutting on species composition.— Clearcutting apparently caused a marked change in the species composition at the Ellicott Rock site. Fifty-five percent of the species in the forest sample are not present in the clearcut sample, and 50 percent of the species in the clearcut sample are not present in the forest sample. The Bray and Curtis (1957) index of similarity, C = 2w/(a + b) [where a = the total number of identified individuals in one sample, b = the total number of identified individuals in the 288 THE JOURNAL OF ARACHNOLOGY Table 2. -Spiders collected at one mature forest site and three recently clearcut sites in the southern Appalachian mountains near Highlands, North Carolina. Stratum designations: G = all speci- mens collected on ground; A = all specimens collected on plants or webs|above ground surface; G, A = majority of specimens collected on ground; A, G = majority of specimens collected above ground. Prey capture mode designations (based upon field observations and literature): H = hunting (cursorial, wandering) spiders; W = web building spiders. Parentheses enclose number of adults collected. Asterisk denotes any species comprising 2.5 percent or more of the total number of individuals collected at that site. CD 3 CD O % o CD £ Oh cd & s OD CD tH u 3 > o U 3 ° x T3 CD O eu £ CD j=i 3 1 cd CD Agelenidae Agelenopsis utahana (Chamb. & I vie) G,A Calymmaria cavicola (Banks) G Grcurina arcuata Keyserling G Grcurina breviaria Bishop & Crosby G Grcurina (immature) G Coras taugynus Chamberlin G Coras (immature) G Cybaeus silicis Barrows G Wadotes carolinus Chamberlin G Wadotes hybridus (Emerton) G Wadotes (immature) G Amaurobiidae Callioplus armipotens (Bishop & Crosby) G Antrodiaetidae Antrodiaetus unicolor (Hentz) G Anyphaenidae Anyphaena pectorosa L. Koch A Wulfila alba (Hentz) A,G Wulfila saltabunda (Hentz) A Araneidae Acacesia hamata (Hentz) A Araneus marmoreus Clerck A Araneus nordmanni (Thorell) A Araneus (immature) A Araniella displicata (Hentz) A Cyclosa turbinata (Walck.) A Leucauge venusta (Walck.) A Mangora placida (Hentz) A Mangora (immature) A Metepeira labyrinthea (Hentz) A Micrathena gracilis (Walck.) A Micrathena mitrata (Hentz) A Neoscona (immature) A Nuctenea cornuta (Clerck) A Araneidae spp. (immature) A Clubionidae Castianeira cingulata (C. L. Koch) G Castianeira longipalpus (Hentz) G Castianeira variata Gertsch G w 8(0) w 1(1) 1(0) 2(2) 2(2) w KD KD w 1(1) 1 w 10(0) 2(0) 20(0) 7(0) w 1(1) w 3(0) 1(0) 2(0) w 5(3) 3(2) 6(2) 9(6) w 1(1) 1(1) 3(3) w 14(14) 3(3) *14(14) *22(22) w 62(0) 37(0) 48(0) 36(0) w *40(8) *17(1; w 3(2) 2(2) 5(2) 4(4) H 1(1) 1(1) H 4(3) H KD W 1(0) 1(0) W 1(0) W 1(1) W 1(0) W 1(1) W 3(1) 5(0) 4(1) W *17(12) 1(1) 4(4) W 4(2) W 1(0) 1(0) W 1(0) 3(0) 1(0) w 4(0) w *15(0) w 2(0) w 3(1) w 2(0) 3(0) H 1(0) H 2(2) KD 4(2) H 1(1) COYLE- EFFECTS OF FOREST CLEARCUTTING ON SPIDERS 289 Table 2.-cont. Taxon ea X ° i •M 2 o W W o £ o ks 3 0) CQ T> in o3 O X o Chiracanthium inclusum (Hentz) Clubiona (immature) Gubionoides excepta (L. Koch) Liocranoides sp. Phrurotimpus alarius (Hentz) Phrurotimpus borealis (Emerton) Scotinella redempta (Gertsch) Scotinella sp. A Scotinella (immature) Clubionidae spp. (immature) Ctenidae Anahita animosa (Walck.) Dictynidae Dictyha sublata (Hentz) Gnaphosidae Cesonia bilineata (Hentz) Drassyllus fallens Chamberlin Drassyllus (immature) Litophillus iemporarius Chamberlin Micaria aurata (Hentz) Micaria longipes Emerton Poecilochroa capulata (Walck.) Zelotes duplex Chamberlin Zelotes hentzi Barrows Zelotes laccus (Barrows) Gnaposidae sp. (immature) Hahniidae Neoantistea agilis (Keyserling) Neoantistea (immature) Hypochilidae Hypochilus thorelli Marx Leptonetidae Leptoneta gertschi Barrows Leptoneta sp. A Leptoneta (immature) Linyphiidae Centromerus denticulatus (Emerton) Ceraticelus carinatus Emerton Ceraticelus fissiceps O. P.-Cambridge Ceraticelus minutus (Emerton) Ceraticelus similis (Banks) Ceratinella brunnea Emerton Ceratinopsidis formosa (Banks) Ceratinopsis interpres (O. P.-Cambridge) Cornicularia directa (O. P.-Cambridge) Erigone autumnalis Emerton H 1(0) H 4(0) 1(0) 3(0) H 2(2) H KD KD H *28(11) *16(11) *30(19) *47(38) H 6(6) 6(5) *13(7) H 1(1) 1(1) 4(4) H 1(1) 4(4) H 4(0) 1(0) 1(0) H 1(0) 2(0) H 13(2) *8(3) W KD 6(3) H 1(0) H KD H 2(0) H 1(0) H 2(0) H KD H KD H 5(4) H KD H KD H 1(0) W KD 1(0) W 1(0) W 1(0) 2(0) W *29(29) *7(7) KD W 5(5) W 45(0) 12(0) 5(0) 7(0) W *34(7) *7(2) *13(2) W 3(3) W *48(42) *14(14) 5(5) *10(10) W 2(2) 8(7) 2(2) W KD W 1(1) W 1(1) W 1(1) *12(12) W 1(1) W 1(1) *13(13) A G,A G G G G G G G G,A G A G G G G A G G G G G G G G A G G G G G A G A G G A,G A G,A 290 THE JOURNAL OF ARACHNOLOGY Table 2.-cont. Taxon Js o CeS o cS to to to > o ex C3 3 U 3 U s o >» to o o p o o O 3 o cj C/5 t-l o s to H Pu o e w £ W to O CQ to 73 o X to 73 Erigone br evident at a Emerton Florinda coccinea (Hentz) Frontinella pyramitela (Walck.) Lepthyphantes zebra (Emerton) Maso sundevalli (Westring) Meioneta unimaculata (Banks) Meioneta sp. A Meioneta sp. B Meioneta sp. C Meioneta (immature) Microneta (immature) Pelecopsidis frontalis (Banks) Pelecopsis moestum (Banks) Pitiohyphantes costatus (Hentz) Scylaceus pallidus (Emerton) Tapinoma bilineta Banks Walkenaera spiralis (Emerton) Linyphiidae sp. A Linyphiidae sp. B Linyphiidae sp. C Linyphiidae sp. D Linyphiidae sp. E Linyphiidae spp. (immature) Lycosidae Lycosa gulosa Walck. Pardosa milvina (Hentz) Pardosa saxatilis (Hentz) Pardosa (immature) Pirata minutus Emerton Pirata montanus Emerton Pirata (immature) Schizocosa ocreata (Hentz) Lycosidae spp. (immature) Oxyopidae Oxyopes salticus Hentz Pisauridae Pisaurina mira (Walck.) Salticidae Agassa cerulea (Walck.) Eris aurantius (Lucas) Eris sp. A Habrocestum pulex (Hentz) Habronattus viridipes (Hentz) Hentzia mitrata (Hentz) Icius elegans (Hentz) Icius sp. A G W A W A W G,A W G W G W G W G W G W G W G W G W G W A W G W G W G W A W A W A W A W G W G,A W G H G H G H G H G H G H G,A H G H G H A H A H A H A H A H G,A H G H A H A H A H 5(5) 1(1) 1(0) 6(3) *12(2) KD 7(2) 3(1) 1(0) *16(0) 1(1) 1(0) 1(1) 2(1) 2(2) KD 1(0) 1(0) KO) 1(1) 1(1) 1(0) 2(2) 1(0) 1(1) 1(1) 1(1) 1(1) 1(1) 22(0) 14(1) 1(1) 1(1) 3(3) 6(0) 25(0) 36(0) *10(2) *13(4) 3(0) 18(16) 1(0) 3(2) KD 3(3) 2(0) 9(4) KD *14(4) 2(0) 1(0) 5(0) 3(0) KD 1(0) 1(0) 1(0) 1(0) 1(0) 1(0) 7(0) 13(5) 5(4) 5(2) 1(0) KD 2(0) 1(0) 1(0) COYLE-EFFECTS OF FOREST CLEARCUTTING ON SPIDERS 291 Table 2.-cont. Taxon Stratum Prey capture mode Ellicott Rock forest Ellicott Rock ; clearcut Buck Creek Clearcut Horse Cove clearcut Maevia inclemens (Walck.) A,G H *34(3) *20(13) *21(5) 8(1) Marpissa lineata (C. L. Koch) G H KD Metaphidippus canadensis (Banks) G H 1(0) 3(0) Metaphidippus flaviceps Kaston A H 1(1) Metaphidippus galathea (Walck.) A H 2(1) 3(3) Neon nellii (Peckham & Peckham) G H 1(1) 2(2) Onondaga lineata (C. L. Koch) A H 1(1) Phidippus prince ps (Peckham) A H 2(2) Phidippus (immature) G H 1(0) Sitticus floridanus Gertsch & Mulaik G H 1(1) Thiodina iniquies (Walck.) A H 4(0) 5(3) 4(3) Zygoballus bettini Peckham A,G H 2(2) 4(3) Salticidae sp. A G H *18(11) *33(21) Salticidae sp. B A H 6(0) 7(0) Salticidae spp. (immature) A H 2(0) Symphytognathidae Mysmena guttata (Banks) G W 2(1) 1(0) 1(0) 5(1) Tetragnathidae Tetragnatha elongata Walck. A W 2(1) 1(1) Tetragnatha seneca Seeley A W 2(1) Tetragnatha versicolor Walck. A W 4(4) Theridiidae Achaearanea rupicola (Emerton) G W 1(1) 1(1) Argyrodes trigonum (Hentz) A W 5(5) 3(3) Dipoena nigra (Emerton) A W 1(1) Episinus amoenus Banks A W 1(1) Euryopis funebris (Hentz) G W 1(1) Pholcomma hirsuta Emerton G,A W *71(21) *16(7) *17(5) 9(5) Robertus frontatus (Banks) G W 14(2) 1(0) Theridion albidum Banks A W 1(1) 1(1) 2(2) Theridion flavonotatum Becker A W 2(2) 2(2) Theridion lyricum Walck. A W 2(2) Thymoites unimaculata (Emerton) A W 1(1) Theridiosomatidae Theridiosoma gemmosa (L. Koch) A,G W 2(1) 2(2) Thomisidae Misumenoides formosipes (Walck.) A H 2(0) 1(0) 3(1) Misumenops oblongus (Keyserling) A H 1(0) 3(1) 1(0) 4(1) Philodromus placidus Banks A H 1(1) Philodromus rufus Walck. A H 1(1) Thanatus sp. (immature) A H 1(0) Xysticus pellax 0. P.-Cambridge G H KD Xysticus (immature) G,A H 4(0) 9(0) 1(0) 2(0) Uloboridae Hyptiotes cavatus (Hentz) A W 4(0) 2(0) 1(0) 292 THE JOURNAL OF ARACHNOLOGY Table 3. -Bray and Curtis similarity indices for the spider samples from all study sites. The Bray and Curtis similarity index is defined in the text. (A) (B) (C) (D) Ellicott Rock forest (A) Ellicott Rock clearcut (B) Buck Creek clearcut (C) Horse Cove clearcut (D) .395 .407 .312 .414 .418 .439 other sample, and w = the sum of the lesser values for those species present in both samples] was used to assess the similarity of species composition among all four sites (Table 3). For the Ellicott Rock sites its value is 0.395, indicating low similarity. The similarity values (Table 3) also indicate that the species composition of each clearcut sample is distinctly different, a result that is consistent with the pre- and post-clearcut vegetational differences among the clearcut sites. Effects of clearcutting on abundance.— Clearcutting reduced the total spider popula- tion at the Ellicott Rock site (Tables 2 and 4). Most of this decrease was in the form of marked declines in nine species which are common in the mature forest: Pholcomma hirusta, Cemticelus fissiceps, Centromerus denticulatus, Leptoneta gertschi, Salticidae sp. A, Leucauge venusta, Micrathena mitrata, Robertus frontatus , and Wadotes hybridus. The only species which appear to have increased markedly as a result of clearcutting are Pardosa milvina, Habrocestum pulex, Phrurotimpus borealis , and Salticidae sp. B. Popula- tions of some species which are common in the forest ( Phrurotimpus alarius, Anahita animosa, Lycosa gulosa , and Maevia inclemens) and some which are less common ( Cybaeus silicis, Antrodiaetus unicolor , and Thiodina iniquies ) apparently were not strongly affected by clearcutting. Effects of clearcutting on species diversity.— Despite its limitations (Peet 1974, 1975) and because of its use in other arachnid studies (Uetz 1975, 1976; Stanton 1979) and the absence of clearly more suitable indices, the Shannon index was used in the present study to measure the species diversity of each sample. The Shannon formula, H' = Jj Pi log „ Pi [where pi = the proportion of total individuals in species i, and s = the number of species] , is a combined measure of both species richness (the number of species) and evenness (the relative equality of species abundance) of a sample. Pielou’s (1966) even- ness index (H'obs/H'max, where H'max = log s) was used, despite shortcomings (Peet 1974, 1975) to measure the evenness component of the diversity index for each sample. The f-test designed by Hutcheson (1970) was used to determine the significance of differences between forest and clearcut sample diversity indices. The species diversity (Table 4) for the total Ellicott Rock forest spider sample is significantly lower than that of both the Ellicott Rock clearcut sample ( t = 2.711, d.f. = 590, P < 0.01) and the Horse Cove clearcut sample ( t - 5.452, d.f. = 701 , P < 0.001) but not significantly lower than that of the Buck Creek clearcut sample ( t = .502, d.f. =654, P > 0.5). The higher diversity, higher equitability (evenness), though lower species richness of the Ellicott Rock clearcut total sample (Table 4) as compared to the Ellicott Rock forest total sample indicates that clearcutting of the Ellicott Rock forest decreased the number of spider species, but made population sizes of the remaining species more equitable (even). COYLE-EFFECTS OF FOREST CLEARCUTTING ON SPIDERS 293 Table 4. -Number of individuals, richness (= number of species in sample), species diversity (= Shannon’s H'), and evenness (= Pielou’s evenness index) of the study site spider samples compared. Ellicott Ellicott Buck Horse Rock Rock Creek Cove forest clearcut clearcut clearcut Total sample: no. individuals 595 297 419 418 richness 60 53 50 71 diversity index 3.1667 3.4198 3.2142 3.6523 evenness .773 .861 .822 .857 Sweep sample: no. individuals 134 46 38 69 richness 22 15 7 21 diversity index 2.1132 2.2272 1.4011 2.6735 evenness .684 .822 .720 .878 Aerial hand s.: no. individuals 45 24 28 39 richness 18 10 12 16 diversity index 2.5421 1.8484 2.2849 2.5374 evenness .880 .803 .920 .915 Ground hand s.: no individuals 30 38 37 14 richness 9 7 13 8 diversity index 1.9405 1.3188 2.3202 1.9351 evenness .883 .678 .905 .931 Pitfall sample: no. individuals 106 111 191 159 richness 16 18 26 28 diversity index 2.3240 2.5952 2.5954 2.5763 evenness .838 .898 .797 .773 Litter sample: no. individuals 280 78 125 137 richness 16 16 14 16 diversity index 1.9914 2.2754 2.0286 2.3390 evenness .718 .821 .769 .844 Effects of clearcutting on different spider guilds. -An examination of the responses of various guilds of spiders to clearcutting may help explain why clearcutting reduced the numbers of individuals and species yet increased the evenness and species diversity in the spider community. Clearcutting at Ellicott Rock had a significant effect upon both the relative abundance (R x C test, P < 0.005) and richness (R x C text, P < 0.05) of the three general guild categories, ground web builders, aerial web builders, and hunting spiders (ground and aerial hunters combined). The net effect of clearcutting at Ellicott Rock was to decrease the numbers and species richness of sedentary ground spiders and aerial web spiders while increasing the species richness of hunting spiders and barely affecting their numbers (Table 5). This conclusion must be tempered by the possibility that the high proportion of hunting spiders in the clearcut sample is due partly to higher pitfall and hand capture rates caused by increased activity of individual spiders in the more variable microclimate of the clearcut site. The data summarized in Table 6 also show that web builder guilds were negatively affected by clearcutting while hunting spiders may have benefited. Eight of the nine common forest species which declined markedly after clearcutting are ground web spiders or aerial web spiders. The only four species which appeared to increase markedly after clearcutting are hunting spiders. In addition, all four of the common forest species which remained abundant after clearcutting are hunting species. 294 THE JOURNAL OF ARACHNOLOGY Table 5. -The guild representation of the study site samples compared. Ellicott Ellicott Buck Horse Rock Rock Creek Cove forest clearcut clearcut clearcut Ground web builders no. (and %) of individuals 334(56) 108(36) 245(59) 185(44) no. (and %) of species 19(32) 14(26) 23(46) 17(24) Ground hunters no. (and %) of individuals 89(15) 105(35) 113(27) 124(30) no. (and %) of species 8(13) 16(30) 12(24) 21(30) Aerial hunters no. (and %) of individuals 50(8) 44(15) 26(6) 52(12) no. (and %) of species 10(17) 12(23) 5(10) 16(23) Aerial web builders no. (and %) of individuals 123(21) 40(14) 35(8) 57(14) no. (and %) of species 23(38) 11(21) 10(20) 16(23) Horse Cove clearcut, like the Ellicott Rock clearcut, has a high proportion of species and individuals of hunting spiders and a low proportion of web building spiders (Table 5). Although the low percentage of species and individuals of hunting spiders and the high proportion of ground web spiders in the Buck Creek clearcut sample appears to weaken the hypothesis that clearcut areas support a larger proportion of hunting spiders and a lower proportion of web builders than forests, it should be pointed out that this site was sampled only a few months after clearcutting. Probably not enough time had passed for the processes of population decline, colonization, and population growth to complete initial major changes in species composition. A comparison of the two Ellicott Rock sites by examining the abundance, richness, diversity, and evenness values of the subsamples collected by each different collecting technique (Table 4) is only marginally helpful in explaining the effects of clearcutting on the spider community, because none of these techniques collects exclusively spiders of a single guild. Although sweeping and aerial hand collecting capture primarily aerial web builders, aerial hunters are also collected. Ground hand collecting yields both ground hunters and ground web builders. Although pitfall traps collect mostly ground hunters, some ground web builders are also collected; the reverse is true for litter samples. Only the aerial hand subsample diversity indices were significantly different ( t = 2.442, d.f. = 43, P < 0.02), with the forest subsample index being higher. Nevertheless, the data for the Ellicott Rock sites’ sweep samples, aerial hand samples, pitfall samples, and litter samples (Table 4) suggest that clearcutting reduces aerial web spider abundance and richness, reduces ground web spider abundance, and does not negatively affect the hunt- ing spider guild. DISCUSSION In summary, the reduction of species richness caused by clearcutting was primarily due to the elimination of certain species of sedentary ground spiders and aerial web spiders, but was moderated by apparent immigration of hunting spider species. Simultaneously, the evenness component of diversity increased by virtue of marked population declines in COYLE-EFFECTS OF FOREST CLEARCUTTING ON SPIDERS 295 Table 6. -Effects of clearcutting at Ellicott Rock on different guilds of spiders. Numbers in paren- theses are precentages of the total number of species in each column. A = species which decreased markedly after clearcutting, B = other species present in forest but not collected in clearcut, C = species which increased markedly after clearcutting, D = other species present in clearcut but not collected in forest. A B C D Ground web builders 5 (56) 7(21) 2(8) Ground hunters 1(11) 2(6) 4 (100) 11(42) Aerial hunters 6(18) 7 (27) Aerial web builders 3 (33) 18 (55) 6(23) most of the dominant forest species (most of which are ground web or aerial web spiders) and by an increase in the abundance of some hunting species. Species diversity increased on the strength of this evenness change. Similar results were obtained by Huhta (1971) in a study of the effects of clearcutting on the ground stratum spiders of spruce forests in Finland. He observed that while clearcutting reduced ground spider abundance, species diversity increased because of an increase in numbers and species of hunting spiders and a concurrent decrease in population densities of previously dominant, sedentary litter species. It seems reasonable to postulate that the apparent decline in the abundance and diversity of ground web builders (nearly all of which live in the litter) following clearcut- ting at Ellicott Rock was due in large part to an increase in the frequency and amplitude of fluctuations of microclimatic factors (especially temperature and humidity) within the litter as a result of 1) the elimination of the forest canopy and its ameliorating effects on ground level microclimate, and 2) the reduction of litter thickness. McGee (1976) found that soil temperatures in a southern Appalachian site during the first summer following clearcutting occasionally exceed 60° C (140°F) at 1/16 inch below the surface and 42° C (110°F) at 1 inch below the surface. Huhta (1971) recorded greater temperature and humidity extremes in the litter of clearcut areas than in simultaneously monitored forest litter and concluded that this microclimate instability was the primary reason for the declining numbers of sedentary litter spiders. Others have demonstrated the moderating effect of sheltering vegetation and/or increased litter depth upon microclimate variation within the litter (Geiger 1950, Gill 1969, Hagstrum 1970). Stanton (1979) concluded that the greater abundance and species diversity of litter mites in forests as opposed to field communities was partly due to a more favorable litter microclimate resulting from the ameliorating effects of both the forest canopy and thicker litter. Other workers have found reduced litter thickness to be correlated with reduced ground spider abundance (Lowrie 1948, Hagstrum 1970, Berry 1967) and diversity and richness (Uetz 1975, 1979). Huhta (1971) has pointed out that a reduction of litter thickness may also reduce spider populations by decreasing the abundance and diversity of intralitter space required for web attachment, refugia, and prey. Uetz (1979) has demonstrated that, at least for ground hunting spiders, decreased litter depth reduces species richness not only because of increased microclimate instability, but also because of reduced litter complexity and reduced prey abundance. The drastic decline in numbers and species of aerial web spiders which utilize the understory vegetation of the forest is probably due primarily to the elimination of the forest canopy and its buffering effects on light intensity, temperature, humidity, and wind (Geiger 1950). Summers (1979), in a recent study of the effects of clearcutting on 296 THE JOURNAL OF ARACHNOLOGY opilionids at Ellicott Rock, has reached a similar conclusion about the primary cause of the drastic reduction in abundance and diversity of opilionids which require forest under- story habitats like those used by aerial web spiders. Some studies show that shortages of appropriate spaces and attachment points for webs probably limit the abundance of some web spiders (Duffy 1962, Cherrett 1964, Judd 1965). However, in view of the abundance of shrubs and young trees in the clearcut study sites, it seems unlikely that a shortage of geometrically suitable web sites was the primary factor limiting aerial web spider success. Cannon (1965) found markedly lower numbers of aerial web spiders in two forests with relatively unstable climates than in an adjacent, climatically more stable forest, even though all three forests appeared to have similar densities of plants suitable for web attachment. Prey abundance can have an important effect on web spider abun- dance (Luczak 1963), but since prey population densities were not measured in the present study, the impact of this factor cannot be estimated. The ability of hunting spiders to successfully contend with, perhaps even take advan- tage of, clearcutting is not surprising in view of the abundant evidence that many hunting spiders are remarkably well adapted to open and climatically harsh environments (Lowrie 1948, Herzog 1961, Turnbull 1966, Schmoller 1970a, 1970b, 1971, Almquist 1973, Riechert and Reeder 1972, Gertsch and Riechert 1976, Uetz 1976). There seem to be two important reasons for the success of hunting spiders in such environments. First, many hunting spiders live on or close to the ground where the climate is relatively stable (Geiger 1950). Secondly, their ability to move readily to patches with more favorable climate and resource values (Williams 1962, Turnbull 1966, Almquist 1973, Kronk and Reichert 1979) may be especially important in enabling them to cope with the large amount of spatial and temporal variation in microclimate that exists in a clearcut habitat. In view of the findings of Uetz (1975, 1979) and Hagstrum (1970) that decreased litter depth is sometimes correlated with decreased species richness and abundance of ground hunting spiders, it is perhaps surprising that ground hunters could increase in richness and abundance following clearcutting. One factor that may compensate for a negative effect of decreased litter depth following clearcutting is an increase in the population densities of some prey species as a result of the high productivity of rapidly regenerating vegetation concentrated close to the ground. Enders (1975) discussed the potentially important influence on spider abundance of both high productivity at inter- mediate stages of succession and the number of prey per volume of habitat space as determined by the vertical distribution of this primary productivity. Another effect of clearcutting which may benefit hunting spiders is the increase of felled decomposing trees and branches which may not only increase the numbers of some decomposer chain arthropods which serve as spider food, but also increase the supply of shelters required by some hunting spiders. ACKNOWLEDGMENTS This research was conducted under a cooperative agreement between the Southeastern Forest Experiment Station of the USDA Forest Service and the Highlands Biological Station, with funds provided by the former and administered by the latter. Drs. Richard Bruce and Stephen Boyce provided helpful advice during the study, and Dr. Bruce saw to it that the facilities of the Station were available when needed. John Horn guided me to the clearcut sites. Dr. J. D. Pittillo analyzed the Ellicott Rock forest vegetation and helped me evaluate its similarity to the adjacent pre-clearcut forest. Drs. Willis Gertsch, COYLE-EFFECTS OF FOREST CLEARCUTTING ON SPIDERS 297 Norman Platnick, and William A. Shear aided in the identification of a few difficult spiders. I wish to thank the following persons for critically reviewing drafts of this paper: Leslie Bishop and Drs. Steven Boyce, Richard Bruce, Roger Lumb, John McCrone, Susan Riechert, and George Uetz. LITERATURE CITED Almquist, S. 1973. Habitat selection by spiders on coastal sand dunes in Scania, Sweden. Ent. Scand., 4:134-154. Berry, J. W. 1967. The distributional ecology of spiders in the old-field succession of the Piedmont region of North Carolina. Unpublished Ph.D. thesis, Duke University. 259 pp. Boyce, S. G. 1977. Management of eastern hardwood forests for multiple benefits (DYNAST-MB). USD A For. Serv. Research Paper SE-168. 116 pp. Bray, J. R. and J. T. Curtis. 1957. An ordination of the upland forest communities of southern Wisconsin. Ecol. Monographs, 27(4): 325-349. Breymeyer, A. 1966. Relations between wandering spiders and other epigeic predatory arthropoda. Ekol. Pol. (A), 14:18-71. Cannon, S. S. 1965. A comparison of the spider fauna of four different plant communities found in Neotoma, a small valley in south central Ohio. Ohio J. Sci., 65(3):97-110. Cherrett, J. M. 1964. The distribution of spiders on the Moor House National Nature Preserve West- moreland. J. Anim. Ecol., 33:27-48. Duffy, E. 1962. A population study of spiders in limestone grassland. The field layer fauna. Oikos, 13:15-34. Enders, F. 1975. The influence of hunting manner on prey size, particularly in spiders with long attack distances (Araneidae, Linyphiidae, and Salticidae). Amer. Natur., 109(970):7 37-763. Geiger, R. 1950. The climate near the ground. Harvard Univ. Press, Cambridge, Mass. Gertsch, W. J. and S. E. Riechert. 1976. The spatial and temporal partitioning of a desert spider community, with descriptions of new species. Amer. Mus. Novitates, (2604): 1-25. Gill, R. W. 1969. Soil microarthropod abundance following old field litter manipulation. Ecology, 50:805-816. Hagstrum, D. W. 1970. Ecological energetics of the spider Tarentula kochi. Ann. Ent. Soc. Amer., 63:1297-1304. Herzog, G. 1961. Zur Okologie der Terrestren Spinnenfauna markischen Kieferheiden. Entom. Zeitschr., 71:1-11. Horn, J. 1976. Vegetational changes after clearcutting in the southern Appalachians. Report to USD A Forest Service, Southeastern Forest Experiment Station, Asheville, N.C. 31 pp. Huhta, V. 1971. Succession in the spider communities of the forest floor after clearcutting and prescribed burning. Ann. Zool. Fennici, 8:483-542. Hutcheson, K. 1970. A test for comparing diversities based on the Shannon formula. J. Theor. Biol., 29:151-154. Judd, W. W. 1965. Studies of the Bryon Bog in southwestern Ontario XVIII. Distribution of Harvest- men and spiders in the bog. Nat. Mus. Canada Natur. Hist. Pap., 28:1-12. Kronk, A. E. and S. E. Riechert. 1979. Parameters affecting the habitat choice of a desert wolf spider, Lycosa santrita Chamberlin and Ivie. J. Arachnol., 7:155-166. Lowrie, D. C. 1948. The ecological succession of spiders of the Chicago area dunes. Ecology, 29:334-351. Luczak, J. 1963. Differences in the structure of communities of web spiders in one type of environ- ment (young pine forest). Ekol. Pol. (A), 11:159-221. McGee, C. E. 1976. Maximum soil temperatures on clearcut forest land in western North Carolina. USDA For. Serv. Research Note SE-237. Moulder, B. C. and D. E. Reichle. 1972. Significance of spider predation in the energy dynamics of forest floor arthropod communities. Ecol. Monographs, 42(4):473-498. Peet, R. K. 1974. The measurement of species diversity. Ann. Rev. Ecol. Sys., 5:285-307. Peet, R. K. 1975. Relative diversity indices. Ecology. 56:496-498. Pielou, E. C. 1966. The measurement of diversity in different types of biological collections. J. Theor. Biol., 13:131-144. 298 THE JOURNAL OF ARACHNOLOGY Riechert, S. E. 1974. Thoughts on the ecological significance of spiders. BioScience, 24(6):352-356. Riechert, S. E. and W. G. Reeder. 1972. Effects of fire on spider distribution in southwestern Wiscon- sin prairies. Proceed. Second Midwest Prairie Conf., pp. 73-90. Schmoller, R. 1970a. Terrestrial desert arthropods: fauna and ecology. Biologist, 52:77-98. Schmoller, R. 1970b. Life histories of alpine tundra arachnida in Colorado. Amer. Midi. Nat., 83:119-133. Schmoller, R. 1971. Habitats and zoogeography of alpine tundra Arachnida and Carabidae (Coleop- tera) in Colorado. Southwest. Nat., 15:319-329. Stanton, N. L. 1979. Patterns of species diversity in temperate and tropical litter mites. Ecology, 60(2):295-304. Summers, G. 1979. The effects of clearcutting on opilionid populations in the southern Appalachians. Report to the USDA Forest Service, Southeastern Forest Experiment Station, N.C. 11 pp. Turnbull, A. L. 1966. A population of spiders and their potential prey in an overgrazed pasture in eastern Ontario. Canadian J. Zool., 44:557-583. Turnbull, A. L. 1973. Ecology of the true spiders (Araneomorphae). Ann. Rev. Ent, 18:305-348. Uetz, G. W. 1975. Temporal and spatial variation in species diversity of wandering spiders (Araneae) in deciduous forest litter. Environ. Ent., 4(5):719-724. Uetz, G. W. 1976. Gradient analysis of spider communities in a streamside forest. Oecologia, 22:373-385. Uetz, G. W. 1979. The influence of variation in litter habitats on spider communities. Oecologia, 40:29-42. Williams, G. 1962. Seasonal and diurnal activity of harvestmen (Phalangida) and spiders (Araneida) in contrasted habitats. J. Anim. Ecol., 31:23-42. Manuscript received May 1978, revised June 1980. Work, R. W. 1981. A comparative study of the supercontraction of major ampullate silk fibers of orb-web-building spiders (Araneae). J. Arachnol., 9:299-308. A COMPARATIVE STUDY OF THE SUPERCONTRACTION OF MAJOR AMPULLATE SILK FIBERS OF ORB-WEB-BUILDING SPIDERS (ARANEAE)1 Robert W. Work Fiber and Polymer Science Program, David Clark Laboratory North Carolina State University, Raleigh, North Carolina, 27607 U.S.A. ABSTRACT An improved and simple technique is described for determining the magnitude of the recently observed phenomenon of supercontraction of major ampullate silk fibers of orb-web-building spiders. Supercontraction ratios obtained by the use of that method and also from an instrumental method are given. It is seen that only three levels of supercontraction are found among twenty two species of orb-web-building spiders. The data are presented on the basis of taxonomic classification and are subjected to statistical analyses in order to determine which ones may be effectively differentiated from the others. A contribution to the integerity of the wetted web is suggested as a function for the retractive driving force which underlies supercontraction. Mention is made of the possible usefulness of supercontraction as a taxonomic symptom. INTRODUCTION This paper is one of a continuing series (Work 1976, 1977a, 1977b, 1978, 1981) which have been concerned with the silks produced by orb-web-building spiders. In the third of these, the discovery was announced of a very large axial shrinkage in water at room temperature of major ampullate silk fibers and its absence in minor ampullate fibers. It had been observed that the former retract to about one half their original lengths when they are wetted in an axially unrestrained condition. In the field of macromolecular chemistry this phenomenon is not unknown, being termed supercontraction. But it was quite unexpected, for although it can be induced in some natural and man-made fibers, it can be produced only under drastic conditions of chemical swelling or elevated tempera- tures, or both. Thus its existence has encouraged an expanded physico-chemical investi- gation, having as its objectives the characterization of the phenomenon and the determi- nation of its basis at the molecular level. It is expected that the fruits of this research will be published in future papers of this series. But since orb webs are commonly wetted by dew and rain, arachnological considerations were not disregarded. Grants from the Research Corporation and the National Science Foundation (currently CPE- 79-08905) supported, in part, the research described in this paper. 300 THE JOURNAL OF ARACHNOLOGY In the earlier part of the study it was found that the magnitude of supercontraction was remarkably uniform among all of the samples examined, as compared with their other physical properties such as axial force-elongation behavior. When one spider was found to have produced silk having an entirely different level of supercontraction, the observation was brought to the attention of Dr. Herbert W. Levi, who indicated that the property might be found to be a taxonomic symptom. Accordingly, while the study has remained centered upon macromolecular chemistry, increased attention was given to taxonomic aspects. To this end, silk samples from as many species of orb -web -building spiders as became available were examined for supercontraction. Techniques for measur- ing this property were improved in order to increase precision. The results which have been secured will be found in this paper. BACKGROUND The silk of Bombyx mori (Linnaeus) has been studied extensively, as is well known. Other silks have received relatively little attention, but workers at the Shirley Institute (Manchester, England) examined a considerable number of them. Lucas, Shaw and Smith (1960a) related the chemical compositions of 68 species of four orders of the class Instecta, and six species of Arachnida, to their taxonomy and also (1955), to a lesser degree, to certain of their physical properties. Associated observations were published by these authors (1958, 1960b) and by Lucas (1964). By their nature, chemical analyses, reported upon as amino acid ratios of the silk polypeptides, are amenable to classification in only the broadest sense. On the other hand, X-ray diffraction measurements of the crystalline domains in most of these same silks allowed Warwicker (1954, 1955, 1960) to classify them. He reported that all had orthorhombic unit cells, which possessed uniform axial and hydrogen bonding dimensions. He placed them in five groups based upon differences in the third, i.e., the side group, dimension. Of the seven arachnid silks, six came from named species. Of the latter, three were from “nests” of species of the family Theraphosidae, and one was a cocoon silk. The remaining two were from orb-web- builders, being “reeled” silk from Nephila madagascariensis (Vinson) and silk of unde- fined origin from Araneus diadematus Clerck. It is possible, even probable, that these last two were in whole or in part from the major ampullate glands. In any case, the two were found to be different and the former was placed in group 3 and the latter in group 5. Other investigators of chemical compositions of spider silks (Braunitzer and Wolff 1955, Fischer and Brander 1960. Peakall 1964, Zemlin 1968, Andersen 1970) have reported their analyses on the basis of species. The same is true of workers (Herzog 1915, DeWilde 1943, Zemlin 1968, Work 1976, 1977b, Denny 1976) who have described physical properties. Since none of these authors has attempted to classify his observations on the basis of taxonomy, it is reasonable to assume that the limitations imposed by the intrinsic natures of such analyses and measurements have precluded this. MATERIALS Fibers obtained from the following species of orb-web-building spiders were examined: Araneus diadematus Clerck, Araneus gemma (McCook), Araneus marmoreus Clerck, Araneus pegnia (Walckenaer), Neoscona hentzi (Keyserling), Neoscona nautica (L. Koch), Verrucosa arenata (Walckenaer), Eriophora fuliginea (C. L. Koch), Nuctenea comuta (Clerck), Nuctenea sclopetaria (Clerck), Argiope argentata (Fabricius), Argiope aurantia WORK- SILK SUPERCONTRACTION 301 Lucas, Argiope trifasciata (Fbrskal), Micrathena gracilis (Walckenaer), mitrata (Hentz), Nephila clavipes (Linnaeus), Nephilengys cruentata (Fabricius), Tetragnatha elongata Walckenaer, Tetragnatha versicolor Walckenaer, Uloborus penicillatus Simon, Uloborus glomosus (Walckenaer), and Hyptiotes cavatus (Hentz). In addition, a few other spiders happened along, Achaearanea tepidariorum (C. L. Koch), Metacyrba undata (DeGeer) and Lycosa rabida Walckenaer, and their trailing silks were sampled. With the exception of one V. arenata, one N. hentzi and the L. rabida , all were females. The primary silk samples from U. penicillatus were taken from webs (W. G. Eberhard) and sent from Colombia. All others were obtained by the writer or his students. METHODS Primary samples of silk fibers have been secured by the forcible silking of large spiders. Where these fibers could be seen exiting from the spiders’ bodies, all from major ampul- late spigots on the anterior spinnerets supercontracted, but none from the minor ampul- late spigots on the median spinnerets has supercontracted. Fibers also were obtained directly from orb webs, either in natural habitat or from caged spiders (Witt 1971). From webs, the loci of sampling were generally the support elements (foundation lines of the first or second order, Kaston 1972: 137: mooring and framework threads, Jackson 1973). A few of the samples were radii or segments of temporary spirals. With captured spiders that did not build webs in cages, samples were secured from the trailing silk left in their movements about the cages, or from induced draglines (Work 1978) or induced trailing silk (Work 1981). The sources of all such silks can only be determined indirectly. In a great many cases, primary samples were found to consist of more than one pair of fibers. Among these, it was often seen that two distinct diameters were represented, their ratio being of the order of two to one. In every such case the larger supercontracted, the smaller did not. Based upon all of this evidence, the criterion for inclusion in the present study has been the presence of axial supercontraction, when wetted at room temperature with distilled water, in an unrestrained condition. It has been inferred that for orb-web- building spiders the sources of these fibers were the major ampullate gland systems. Hereinafter, unless otherwise described, the word “fiber(s)” refers only to those assumed to originate in the major ampullate gland system. Induced trailing silks of the non-orb- building spiders were examined essentially out of curiosity and the results have been included for such interest as they may provoke. The methods of obtaining samples, examining them for imperfections, manipulating them, and making the measurements, have been described in earlier papers (Work 1976, 1977b and 1981). Added experience has led to modifications that future investigators may find to be useful. In recent stages of the study, all samples have been taken directly onto glass microscope slides (plastic causes problems with static), to which they are fastened by means of small tabs of single faced self-adhesive tape. This provides for easy microscopic examination by transmitted light at 250-430X. For wider study of the phys- ical properties of these silks, it has been necessary to reject many due to the presence of discontinuities or flaws, cemented sections, or because the sample consisted of multiple pairs which could not be easily separated. But when the only objective is to determine the magnitude of supercontraction, a greater incidence of useful samples may be expected. A sample consisting of a single pair is to be preferred; two pairs may be useful; a complex of more than two usually defies manipulation. Experience has shown that when the sample is composed of four fibers of approximately the same diameter, these are more often than 302 THE JOURNAL OF ARACHNOLOGY not in line contact for their entire lengths. With this situation only rarely can they be separated successfully into two component pairs, with subsequent transfer of one pair to a second glass microscope slide, without incurring the hazard of causing damage which may invalidate measurements. On the other hand, if two of the four are about one half the diameter of the other two, it may be assumed that the sample consists of a pair of major and a pair of minor ampullate fibers. In such a case, it is often seen that the two pairs do not remain entirely in line contact, since the smaller pair tends to loop slightly away from the larger pair. Such loops provide vantage points for the insertion of a micro dissection needle (Clay-Adams) between the pairs for separation, with one pair then being transferred to another slide for individual study. There is always the possibility that the original sample is other than straight, or more often, in a state of slight strain, similar to the situation in which it existed when it was in the web. To correct either condition, the tab at one end is lifted and while illuminated at a glancing angle, perpendicular to the axis of the sample, it is allowed to form a slight catenary which is then brought to linearity and refastened to the slide. When it is a catenary it is axially unrestrained, and during these moments it is vital that the experi- menter, necessarily close to it, should not breathe on it since condensation of moisture on it will initiate supercontraction. Although this investigator has not examined the major ampullate fibers for sensitivity to high relative humidities, it is suggested that these materials should not be manipulated under moist atmospheric conditions. Subsamples for measurement of supercontraction (if any) are prepared by cementing the sample to the microscope slide at intervals of 5-10 mm. The electrical conducting “paint”, commonly used for attaching specimens to studs for scanning electron micro- scopy is useful for the purpose and is easily applied with a double zero camel’s hair brush. This cement is “quick drying,” forms a strong adhesive bond to both glass and silk, and being observable by the unaided human eye, makes obvious the location of the subsam- ples. Also, it has less tendency to run into the sample adjacent to the cemented spot than other equally adhesive cements, such as catalyzed epoxy materials. Since the original publication of the method of determining supercontraction (Work 1977b: 653) an improvement in the technique has been developed. As before, the length of the subsample is first measured. Instead of then immersing it completely in a droplet of water, followed by cutting it free of the cement at one end, the reverse is done. This makes it possible to place the droplet of water adjacent to the freed end and move it with a camel’s hair brush until it touches the end. Capillarity then draws the water progres- sively along the fibers, which supercontract upon contact. This tends to prevent a snap- ping back which may occur at the instant of cutting of a completely immersed subsample. With the suggested improvement, the fibers shrink and retract axially without excessive lateral motion. This modification is particularly important when the subsample consists of a pair each of major and minor ampullate fibers. With the original method, the violent retraction of the former fibers usually caused them to entangle the non-shrinking latter pair. With the improved technique, the slowly retracting major pair tends in many in- stances to draw away from the dimensionally stable minor pair. In the sections to follow, this entire procedure will be referred to as the “slide” method, because the sample is cemented to a microscope slide, and the measurement obtained will be termed as “slide supercontraction,” this being the ratio of the supercontracted length to the original length. The method of producing and measuring supercontraction through the usage of the Instron® tensile tester has been described (Work 1977b: 654) and has continued to be WORK- SILK SUPERCONTRACTION 303 essential in connection with the broad study of this phenomenon and the force- elongation behavior of the supercontracted fibers. The measurement so secured will be referred to in what follows as “Instron supercontraction,” being the ratio of fully retract- ed length to original length. The data obtained in this investigation have been subjected to statistical analyses, the results of which are presented in the next section. The methods used and the computer programs for them were developed by the staff of the SAS Institute (1979). Computa- tions were made by means of an IBM® 370/165 computer at the Triangle Universities Computer Center. RESULTS Table 1 presents the results obtained for the following groups. (A) The slide and Instron supercontraction of major ampullate fibers from 22 species of orb -web -building spiders. (B) The slide supercontraction of the forcible trailing silk fibers of A. tepidar- iorum, a spider which does not build an orb web. (C) The small shrinkage of forcible trailing silk fibers from L. rabida and M. undata , which do not build orb webs. The mean values from these two are compared with the mean value of minor ampullate fibers of eight of the species found in group A. For the samples from each species the data include N, the number of spiders from which samples were secured, N, the number of observations used in making the calcula- tion of the mean value, the mean value, and the coefficient of variation for each such calculation. These data are presented on a taxonomic basis provided by Levi (per. comm.). The cells (species) of group A, within each of the two sets, Slide Supercontrac- tion and Instron Supercontraction, are neither symmetrical from the standpoint of spiders involved, nor uniform in size, as would be desirable for statistical analyses. If Student’s t tests were to be used to determine whether each mean value in a set differs significantly from every other mean value in it, this would require 231 and 136 compari- sons for the set Slide Supercontraction and the set Instron Supercontraction, respectively. Duncan’s New Multiple-Range Test, described in many tests, as for example by Steel and Torrie (1980), simplifies the problem of analysis. The raw data upon which the mean values of supercontracting fibers of Table 1 are based were subjected to an analysis of variance to establish overall species differences, (F = 22.08, P < 0.0001 for set Slide Supercontraction and F = 9.55, P < 0.0001 for set Instron Supercontraction). The Duncan procedure provides each mean value within a set with one or more alphabetic letters. When the same letter is given to mean values within a set, it is not possible to effectively differentiate between them, using the Duncan analysis. The calculations made for the set Slide Supercontraction are independent of those for the set Instron Supercon- traction and the Duncan letters found in one must not be associated with those of the other. DISCUSSION It will be seen in Table 1 that the individual mean values of slide supercontraction ratios are generally slightly higher than those for the comparable Instron supercontrac- tion. The means are 0.564 and 0.543 respectively, for all samples from the subfamily Araneinae. In order to further examine this difference, the data on those samples in this set upon which both slide and Instron supercontraction ratios had been determined on 304 THE JOURNAL OF ARACHNOLOGY adjacent subsamples (N = 43) were subject to a f-test. It was found that the difference between the means of the two variables gave a f -value of 2.202 with a significance probability level of 0.030. This suggests that the two means came from different popula- tions, which in turn implies that the two methods of measuring supercontraction provide a small, but real, systematic difference. It is possible to suggest an explanation of both the difference, as such, and its direc- tion. With the Instron device a sample several centimeters long is wetted, allowed to supercontract to its maximum level, and dried, at which condition it possesses a very high modulus. This being the case, when it is straightened and stretching starts, there is an immediate response by the recording pen. Reversal of the motion of the crosshead drops the strain level to zero. But with the slide technique the supercontracted fiber must be made straight while wetted, at which time it has a very low modulus. Neither the sensi- tivity of the human eye nor the hand manipulating a camel’s hair brush is great enough to differentiate between straightness at zero strain and stretching to a small degree. For example, the difference between the two mean ratios mentioned above, 0.564 and 0.543 is 0.021, which is equivalent to only about 50-100 [im of strain in a sample originally a few mm long. It follows that the Instron provides more accurate data and based upon the lower coefficients of variation in set Instron Supercontraction, seen in Table 1, these are more precise. But when consideration is given to the cost of the Instron device, its general unavailability in biological laboratories, the considerable delicacy of manipulation in- volved in the preparation of the sample and the comparatively large consumption of time required to secure a single datum, the practical value of the slide supercontraction tech- nique is apparent. There are similarities in the formation of major ampullate and strong man-made fibers (Work 1977a) and in their properties (Work 1976). (Biologists who are not familiar with the production and properties of the latter materials may wish to refer to Work (1974) for a summary, or to the Encyclopedia of Polymer Science and Technology for more comprehensive information). But supercontraction brings out distinct differences be- tween man-made and major ampullate fibers. The former may supercontract under drastic conditions of chemical treatment or temperature or both. The amount of shrinkage is a function of amount, speed, and temperature of stretching (technically known as drawing) during manufacture, and the conditions of treatment used to produce supercontraction. On the other hand, whether the latter fibers are produced rapidly as draglines or more slowly as trailing silks, or by pulling out with the fourth leg, the level of supercontraction for each species is uniform, and occurs in water at room temperature, a condition com- monly encountered. These differences suggest a function for the driving force which triggers supercontrac- tion, without regard for the three distinct supercontraction ratios, about 0.55, 0.63, and 0.82, which may be of taxonomic significance. It is, of course, common knowledge that an undisturbed “dry” orb web is in a state of static tension, a condition considered by Denny (1976) and by Wainwright (1976, esp. Fig. 8.2 and associated text). The axial retractive forces of the major ampullate fibers, as radials and supportive elements, are not known, nor are the correlative levels of strain. But it is reasonable to assume that the state of these is within the so-called Hookean, essentially linear and elastic region of the force-elongation response of these fibers, before the yield section is reached. The Hookean region and its attendant yield section disappear when major ampullate fibers are wetted (Work 1977b, Fig. 1), a condition to be expected of fibers which are highly sorbative of and swollen by water. In such a state, the weight of WORK- SILK SUPERCONTRACTION 305 Table l.-The shrinkage of certain spider fibers when axially unrestrained and wetted with water at room temperatures. Explanations of the column headings are given in the text. Classifications by Levi (per. comm.). Supercontraction Slide Instron N N N N Genus Species sp. obs. Mean Cv. % Duncan sp. obs. Mean Cv. % Duncan Araneidae: Araeinae Araneus diadematus 18 41 0.552 13.6 A 4 5 0.570 5.1 A Araneus gemma 1 9 0.578 8.2 ABC 1 3 0.560 1.0 A Araneus marmoreus 3 8 0.520 12.1 A 2 4 0.492 6.7 A Araneus pegnia 1 4 0.560 3.1 ABC 1 3 0.553 6.8 A Neoscona hentzi 6 24 0.545 11.0 A 4 7 0.529 10.3 A Neoscona nautica 1 2 0.605 1.2 ABC - - - - - Verrucosa arenata 3 6 0.548 14.3 A 2 4 0.500 2.8 A Eriophora fuliginea 2 2 0.540 0.0 A 2 2 0.595 10.7 AB Nuctenea cornuta 2 10 0.563 10.2 AB 1 4 0.535 12.4 A Nuctenea sclopetaria 3 8 0.561 12.8 AB - - - - - Araneidae: Argiopinae Argio pe argen ta ta 2 5 0.572 11.6 ABC 1 1 0.570 _ AB Argio pe a uran tia 5 11 0.548 9.0 A 5 7 0.544 7.3 A Argio pe trifasciata 2 7 0.586 13.0 ABC 1 3 0.553 5.5 A Araneidae: Gasteracanthinae Micrathena gracilis 4 18 0.636 10.4 C 2 5 0.590 15.9 AB Micrathena mitrata 1 6 0.560 13.2 AB 1 2 0.535 4.0 A Araneidae: Tetragnathinae Tetragnatha elongata 1 3 0.510 22.1 A 1 3 0.553 11.1 A Te tragna tha versi color Araneidae: Nephilinae 2 3 0.537 2.2 A 2 3 0.553 12.0 A Nephila clavipes 6 22 0.635 11.5 C 5 7 0.650 11.0 B Nephilengys cruentata Uloboridae 3 17 0.822 3.6 D 3 5 0.800 12.7 C Uloborus glomosus 3 8 0.832 4.4 D Uloborus penicillatus 7 7 0.810 7.4 D Hyptiotes cavatus Theridiidae 2 14 0.619 10.9 BC Achaearanea tepidariorum 1 10 0.583 9.7 ABC Mean ratio, final/orig. Cv. % Lycosidae Lycosa rabida Salticidae 1 4 0.960 2.3 Metacyba undata 1 3 0.987 0.6 Various minor ampullate 8 35 0.956 4.0 droplets of water on the elements of the orb web, place them under greater axial stress than in the unwetted static state. It follows that the increased weight will tend to extend the fibers, but this will be opposed by the newly developed retractive forces, which, if they were to be axially unrestrained, will cause them to supercontract. To what degree this retractive force may support the added load of water cannot be estimated at this time. 306 THE JOURNAL OF ARACHNOLOGY It is perhaps fortunate that casual attention was given to three spiders that do not build orb webs, because the trailing silk of A. tepidariorum supercontracts to a level which does not allow it to be effectively differentiated from fibers of the subfamilies Araneinae, Argiopinae, and Tetragnathinae of the family Araneidae. It must be concluded that whatever may be the function, it is not limited to contributing only to the integrity of a wetted orb web. This raises the question as to whether the major ampullate silk of species which build orb webs only in extremely dry habitats supercontract and also whether supercontracting fibers are produced only in climates where they may be expect- ed to be exposed to wet conditions. Only one datum can be added in this connection. Fibers taken from the web of a TV. cruentata in the hot dry climate of the Taita Hills, Kenya, East Africa, in no way differed from those produced by the same spider in the moist climate of North Carolina. When attempting to secure observations on silks from as many species as possible, recourse was made to samples secured in the very early stages of the overall program, which resulted in a rather interesting peripheral observation. A few samples which origi- nally had been considered to be too complex for manipulation had been stored in closed microscope slide boxes, and thus, most importantly, had not been exposed to light, which may be expected to cause degradation of organic fibers. Two samples from E. fuliginea were 197 and 230 weeks old; two from A. argentata , 173 weeks; three from A. aurantia, 206, 219, and 227 weeks; one from A . gemma , 119 weeks. All supercontracted in the same manner as new samples and the data secured from them are included in Table 1 . Under the conditions of storage it is apparent that the dormant or potential retractive stresses do not decay during a period of several years. On the other hand, when, as a matter of curiosity, force-elongation measurements were made on these same fibers, it was found that they ruptured at about one third to one half of the values expected for new fibers. Considering the circadian life and ephemeral nature of the orb web, the resistance of one of its component silks to aging is noteworthy. In taking up the subject of supercontraction as a possible taxonomic symptom, it is well to first examine the data seen in Table 1 for Af. gracilis and M. mitrata. The Duncan letters assigned to them indicate that based upon the slide technique they can be differen- tiated but the Instron data shows that this is not possible. In an attempt to resolve this uncertainty the raw data were subjected to /--tests. In the set Slide Supercontraction a t value of 2.222 and a significance probability level of 0.058 were secured; for set Instron Supercontraction the corresponding figures were 1.238 and 0.274. It must be concluded that the two species produce fibers which do not differ significantly. But more impor- tantly, these findings emphasize that additional measurements must be made so as to make definitive conclusions possible. Similarly, it would have been advantageous if a greater number of both spiders and samples had been available for some of the species listed in Table 1 . This would have reduced the uncertainties which exist where a species carries more than one Duncan letter. But it is quite clear that fibers from the several species and genera listed in Table 1, of the subfamilies Araneinae, Argiopinae, and Tetra- gnathinae of the family Araneidae, cannot be effectively differentiated. The two related species (Levi, per. comm.) in the subfamily Nephilinae are differentiated from this group and from each other. In connection with this former situation it is well to be reminded, as earlier noted, that Warwicker (1960) placed A. diadematus and TV. madagascariensis in different groups, based upon X-ray diffraction diagrams. That the supercontraction ratio of about 0.82 for TV. cruentata is not unique to one species is clear, when it is seen that two uloborids produced the same level. The supercontraction of the fibers from the other WORK- SILK SUPERCONTRACTION 307 uloborid, H. cavatus, resembles that of fibers from N. clavipes. There is little that can be said regarding the lack of supercontraction for the trailing silks of L. rabida and M. undata. The grandular sources of these are unknown. They are dimensionally stable axially when unrestrained in water, as are minor ampullate fibers. The small, reversible, axial, shrinkages exhibited by these fibers are not unexpected among fibers which are highly swollen laterally by water, a property of minor ampullate fibers recorded by Work (1977b, Table 1). Thus the data seen in Table 1 illustrate that there are three distinct levels of supercon- traction. Although I do not have equipment sufficiently sensitive to discriminate between the driving forces which produce these, it is reasonable to assume that corresponding differences exist and are based fundamentally upon molecular structure. Furthermore, it is seen that these are not transient factors. This combination strengthens the suggestion that the entire phenomenon is basic, and as such, may be a taxonomic symptom. The data summarized in Table 1 may be of interest to students of taxonomy, even as the phenomenon itself has stimulated its study from the standpoint of macromole cular chemistry. The measurement is easily performed through the use of commonly available equipment, on such spider silks as any person may come upon. I wish to emphasize that although I happen to be the first one to observe supercontraction in spider silk, I also urge other investigators, whose imaginations may be aroused by it, to expand upon this very small beginning. ACKNOWLEDGMENTS Dr. Herbert W. Levi provided the motivation for this study, gave counsel during its progress, and identified many of the spiders used. Dr. William G. Eberhard supplied primary silk samples. Drs. John F. Andersen, Cesar Edes, Cynthia Grant, Donald Kimmel, Brent Opell (who also identified uloborids), Michael E. Robinson and Messrs. H. H. Newton, W. A. Nattleton and Jesse Perry supplied spiders. Dr. Robert J. Monroe contri- buted guidance concerning statistical analyses. My wife, Anne, has borne with my re- search activities since my official retirement several years ago, helped with the forcible silking of spiders, and edited and typed this paper. To all of these, and to others who helped along the way, I express my sincere appreciation. LITERATURE CITED Andersen, S. O. 1970. Amino acid composition of spider silks, Comp. Biochem. Physiol., 35: 705-711. Braunitzer, G. and D. Wolff 1955. Vergleichende chemische Untersuchungen iiber die Fibroine von Bombyx mori and Nephila madagascariensis. Zeitschriff fur Naturforschung, Series BIO: 404-408. Denny, M. 1976. The physical properties of spider’s silk and their role in the design of orb webs. J. Exp. Biol., 65: 483-506. DeWilde, J. 1943. Some physical properties of the spinning threads of Aranea diadema L., Arach. Neerl. Physiol., 27: 117-132. Encyclopedia of Polymer Science and Technology. 16 vol. and 2 supp. (N. M. Bikales, exec, ed.) 1964-1977, Interscience Publishers, New York. Fishcer, F. G. and J. Brander 1960. Eine Analyse der Gespinste der Kreuzspinne. Hoppe-Seyler’s Z. Physiol. Chem., 320: 92-102. Herzog, A. 1915. Zur Kenntnis der Spinnenseide. Kunststoffe, 5: 25-27, 53-55. Jackson, R. R. 1973. Nomenclature for orb web thread connections. Bull. British Arachnol. Soc., 2(7): 125-126. 308 THE JOURNAL OF ARACHNOLOGY Kaston, B. J. 1972. How to know the spiders. 2nd ed. Wm. C. Brown Co., Dubuque, Iowa, USA. 289 pp. Lucas, F., J. T. B. Shaw and S. G. Smith 1955. V- The chemical constitution of some silk fibroins and its bearing on their physical properties, J. Tex. Inst., 46: T440-T452. Lucas, F., J. T. B. Shaw and S. G. Smith 1958. The silk fibroins. Pp. 107-242, In Vol. 13, Advances in protein chemistry, (C. B. Anfinsen, Jr., Ed.). Academic Press Inc. New York, 525 pp. Lucas, F., J. T. B. Shaw and S. G. Smith 1960a. Comparative studies of fibroins. I The amino acid composition of various fibroins and its significance in relation to their crystal structure and taxon- omy. J. Mol. Biol., 2: 339-349. Lucas, F., J. T. B. Shaw and S. G. Smith 1960b. The composition of arthropod silk fibroins. XI Int. Kongr. f. Entom., Wien 1960. Verh. Bill, 208-214, Verlag. Inst. Ent. Univ., Pavia. Lucas, F. 1964. Spiders and their silks. Discovery, 25:2-26. Peakall, D. B. 1964. Composition, function and glandular origin of the silk fibroins of the spider Araneus diadematus Cl., J. Exp. Zool., 156: 345-352. SAS Staff 1979. SAS user’s guide, 1979 ed., SAS Institute Inc., P.O. Box 10066, Raleight, North Carolina 27605, 494 pp. Steel, R. G. D. and J. H. Torrie. 1980. Principles and procedures of statistics, 2nd ed. McGraw-Hill Book Co., New York. 633 pp. Wainwright, S. A. 1976. Principles of the strength of materials: phenomenological description. Pp. 6-44, 35 3-354. In Mechanical design in organisms. (S. A. Wainwright, W. D. Biggs, J. D. Currey, and J. M. Gosline, eds.) John Wiley & Sons, New York, 423 pp. Warwicker, J. O. 1954. The crystal structure of silk fibroin. Acta. Crystallogr., 7: 565-573. Warwicker, J. O. 1956. The crystal structure of silk fibroins. Faraday Soc. Trans., 52: 554-557. Warwicker, J. O. 1960. Comparative studies of fibroins. II The crystal structure of various fibroins. J. Mol. Biol., 2: 350-362. Witt, P. N. 1971. Instructions for working with web-building spiders in the laboratory. Bioscience, 21 : 23-25. Work, R. W. 1974. Man-made textile fibers. Pp. 301-343. In Riegel’s handbook of industrial chemistry, 7th ed. (J. A. Kent, ed.) Van Nostrand Reinhold, New York, 802 pp. Work, R. W. 1976. The force-elongation behavior of web fibers and silks forcibly obtained from orb-web-spinning spiders. Text. Res. J., 46: 485-492. Work, R. W. 1977a. Mechanisms of major ampullate silk fiber formation by orb-web-spinning spiders. Trans. Amer. Micros. Soc., 96: 170-189. Work, R. W. 1977b. Dimensions, birefringences, and force-elongation behavior of major and minor ampulate silk fibers from orb-web-spinning spiders-The effects of wetting on these properties. Text. Res. J., 47: 650-662. Work, R. W. 1978. Mechanism for the deceleration and support of spiders on draglines. Trans. Amer. Micros. Soc., 97: 180-191. Work, R. W. 1981. Web components associated with the major ampullate silk fibers of orb-web- building spiders. Trans. Amer. Micros. Soc., 100(1): 1-20. Zemlin, J. C. 1968. A study of the mechanical behavior of spider silks. Technical report 69-29-CM, AD 684333, U. S. Army Natick Laboratories, Natick, Massachusetts 01760, USA. Manuscript received May 1980, revised August 1980. Cokendolpher, J. C. 1981. The harvestman genus Liopilio Schenkel (Opiliones, Phalangiidae). J. Arachnol., 9:309-316. THE HARVESTMAN GENUS LIOPILIO SCHENKEL (OPILIONES: PHALANGIIDAE) James C. Cokendolpher Department of Biological Sciences Texas Tech University, Lubbock, Texas 79409 ABSTRACT The harvestman genus Liopilio Schenkel is revised. Its relationship to other harvestmen is briefly discussed. The male of Liopilio glaber Schenkel is described and illustrated for the first time. Liopilio yukon, n. sp., is described from Alaska and the Yukon Territory. INTRODUCTION The original description (Schenkel 1951) of Liopilio is vague. Schenkel’s description could easily serve for immatures of many of the known species of Phalangiidae. Coken- dolpher (1980) noted the similarity of Liopilio to Leptobunus Banks, but did not state to which family these genera belonged. Liopilio and Leptobunus are members of the Phalan- giidae and share several characters with Mitopus Thorell and Tchapinius Roewer. Follow- ing the classical system of Roewer (1923), Leptobunus and Tchapinius are members of the Leptobuninae (Leptobunidae of Rambla 1977, Star^ga 1978; and Leiobunidae of Silhavy 1960), Mitopus is a member of the Oligolophinae (Phalangiidae), and Liopilio as described by Schenkel is a member of the Phalangiinae (Phalangiidae). Recent studies (in preparation) of the “Leptobuninae” reveal that this group is polyphyletic and that Lepto- bunus and Tchapinius are related to the “Oligolophinae.” The single character used to separate the Oligolophinae from the Phalangiinae is the presence or absence of a ventral tooth on the basal segments of the chelicerae. Mitopus, Leptobunus , and Tchapinius have well developed teeth. In Liopilio the teeth not only vary in size due to age and sex of the specimen, but may be entirely lacking. Similarly, the Old World genus Paroligolophus Lohmander has one species, P. agrestis (Meade), which has a cheliceral tooth, and another species, P. meadii (Pickard-Cambridge), which lacks the structure. With such variability, it seems unsound to maintain the Oligolophinae separate from the Phalangiinae. As Liopilio is inadequately described and as it has never been diagnosed or compared to other harvestmen genera, I now present a redescription of the genus and the single described species. In addition, a new species will be described. 310 THE JOURNAL OF ARACHNOLOGY MATERIALS AND METHODS Acronyms for collections from which specimens were examined are listed in the acknowledgments. Specimens in my personal collection are listed JCC. All anatomical measurements are in millimeters and were made using a binocular microscope equipped with an ocular micrometer as outlined by Cokendolpher (1981). Descriptions are based on all available specimens. Measurements were taken on all speci- mens, but due to the small sample sizes only the ranges are reported. Genus Liopilio Schenkel Liopilio Schenkel 195 1:5 1; Forcart 1961:53; Cokendolpher 1980:134. Type species.— Liopilio glaber Schenkel, by monotypy. Diagnosis and comparisons.— Liopilio is similar to Leptobunus, Mitopus, and Tchapi- nius. The four genera differ from other phalangiid genera by lacking strong spines or tubercles on the preocular area, by having the supracheliceral lamellae smooth and short, palps without apophyses, and by generally having a tooth on the basal segment of the Figs. 1-11. -Anatomy of Liopilio : 1, L. glaber, dorsal aspect of male body; 2, L. yukon, dorsal aspect of male body; 3, L. yukon, lateral aspect of immature chelicera; 4 ,L. yukon, lateral aspect of male chelicera; 5, L. yukon, lateral aspect of female chelicera; 6, L. yukon, lateral aspect of male palp; 7, L. yukon, ventral aspect of male palp (excluding femur) with setae removed; 8, L. yukon, medial aspect of male palp; 9, L. yukon, medial aspect of female palp; 10, L . yukon, dorsal aspect of female palp (excluding femur); 11 , L. yukon, lateral aspect of female palp. COKENDOLPHER-THE HARVESTMAN GENUS LI O PI LI O 311 chelicerae. Liopilio and Leptobunus differ from Mitopus and Tchapinius by having the glans of the penis convex beneath, lack of large spines or tubercles anywhere on the body (except for coxae and bases of femora), and by having pseudosegments in tibiae II and all metatarsi. Liopilio differs from Leptobunus by having the palpal femora and patellae expanded distally and by having the male palpal tarsi toothed ventrally. The paired primary setae of the penis glans are located near the stylus junction; whereas in Lepto- bunus they are removed a considerable distance to about the midpoint of the glans. Liopilio is also unusual in the possession of yellow and green pigments on the dorsum of the abdomen. Description.— Medium sized phalangiids with soft, non-tuberculate bodies (Figs. 1, 2); dorsum and venter with only scattered setae. Ocular tubercle low, canaliculate, slightly wider than long, with scattered setae. Chelicerae not enlarged, with or without ventral tooth on basal joint; movable finger without apophysis (Figs. 3-5). Supracheliceral lamel- lae short and smooth. Scent gland pores visible from above, elongate, length slightly less than diameter of eye. Palpi (Figs. 6-11) with inner distal margins of femora and patellae slightly expanded, covered only with setae; distal ends of femora with series of slit sensilla on dorsolateral margins, dorsomedial campaniform sensilla lacking; tarsi and sometimes tibiae and patellae of males with ventral rows of denticles, unarmed in females; claws toothed. Legs relatively short; all articles round in cross-section; femora I less than body length; tibiae II with 2-5 pseudosegments; all metatarsi with 1-5 pseudosegments; trochan- ters and bases of femora sometimes with few black tubercles, otherwise only with setae. Penis as in Figs. 12-15; truncus not grooved on inner distal margin; glans round with primary setae distally, stylus long, much longer than glans setae; stylus-glans membrane dorsally expanded. Ovipositor as in Figs. 16, 17, with 1-2 slit sensilla per side on second segment. Seminal receptacles consisting of paired loops (Figs. 18, 19). Subordinate taxa —Liopilio contains but two species, L. glaber Schenkel and L. yukon , n. sp. Distribution. -Alaska in the United States and Yukon Territory, British Columbia, and Alberta in Canada (Fig. 20). Natural history.— Liopilio spp. have generally been collected at altitudes from 1060-3000 m. Specimens from northwestern Alaska though, are only known from two collections at or near sea level. The exact location of one of these collections is uncertain. These specimens are part of the Vega Expedition (1879) collection and are labeled Port Clarence, Alaska. Holm (1970) reports this collection was made at “Grantley Harbour and Iman suk.” Both adults and immatures of Liopilio glaber have been collected from mid July to early September. Females during this period are filled with eggs. Immatures of Liopilio yukon have been collected during June, early July, and early September. A single immature was collected on 22 March. Adults of L. yukon are only known from specimens collected during mid July and early September. Females at these times are filled with eggs. No other natural history data are available. Liopilio glaber Schenkel Figs. 1, 12, 13, 16, 18,20 Liopilio glaber Schenkel 1951:51, fig. 48; Forcart 1961:53; Cokendo lpher 1980:134. Types.— Female lectotype and immature from Canmore (Casa’de Mountain ?), Banff National Park, Alberta, 3-4 September 1939 (H. Schenkel-Rudin), NMB no. 91-a, exam- ined. 312 THE JOURNAL OF ARACHNOLOGY Distribution.-British Columbia and Alberta in Canada (Fig. 20). Description.— Males: Total length 3.45-4.00, greatest width 2.20-2.60, maximum height 1.62-2.30; cephalothorax yellow brown with dark brown splotches; abdomen and posterior rim of cephalothorax light olive green to gray green; dorsum of abdomen with faint dorsal pattern of dark olive green, interrupted with rows of dark brown spots and few scattered white spots (Fig. 1). The green pigments of the abdomen appear to fade in alcohol to gray or light brown. Ocular tubercle length 0.35-0.38, width 0.40-0.45, height 0.17-0.20, distance from anterior edge of carapace 0.30-0.31 ; yellow brown rings around eyes. Coxae, genital operculum, and abdominal sternites light brown to yellow brown; sternites with lateral margins gray, sternite junctions with rows of dark spots. Chelicerae yellow brown with brown splotches laterally, teeth black, basal segments of each chelic- era usually with a small tooth or ridge ventrally. Palpi light brown to yellow brown with dark brown stripes on dorsa of femora, patellae, and proximal 3A of tibiae; distal tips of Figs. 12- 19. -Genitalia of Liopilio : 12, L. glaber, penis; 13, L. glaber, distal end of penis; 14, L. yukon, penis; 15, L. yukon, distal end of penis; 16, L. glaber, ovipositor; 17, L. yukon, ovipositor; 18, L. glaber, seminal receptacle; 19, L. yukon, seminal receptacle. Scale lines for Figs. 12, 14 = 0.5 mm, 13, 15 = 0.09 mm, 16, 17 = 0.25 mm, 18, 19 = 0.05 mm; BL = basal lateral loops of seminal receptacle, G = penis glans, PS = paired primary setae of glans, S = penis stylus, SS = slit sensilla of second ovipositor segment. COKENDOLPHER-THE HARVESTMAN GENUS LIOPILIO 313 tarsi dark brown; ventral surtaces of tarsi with many dark brown to black denticles arranged in two rows. Palpal lengths: femora 0.98-1.13, patellae 0.55-0.63, tibiae 0.75-0.80, tarsi 1 .40-1 .70. Legs yellow brown with dark brown bands preceded by yellow brown on distal ends of femora I, III, IV; femora II, patellae, and tibiae often with subdistal bands of dark brown; tibiae II with 3, rarely 4, pseudosegments. Femora I-IV lengths (respectively): 3.00-3.43, 6.60-6.93, 3.50-3.87, 4.40-4.69. Tibiae I-IV lengths (respectively): 2.80-3.20, 6.09-6.43, 3.00-3.01 , 3.23-3.38. Penis as in Figs. 12, 13. Females: Form and coloration essentially as in males, except dorsum of abdomen with more green coloration, bases of coxae splotched with brown. Basal segments of chelicerae at most with rounded lobes beneath. Total length 5.31-6.58, greatest width 3.75-4.63, maximum height 2.63-3.40. Ocular tubercle length 0.38-0.43, width 0.42-0.50, height 0.18-0.24, distance from anterior edge of carapace 0.28-0.38. Palpal lengths: femora 0.99-1.39, patellae 0.52-0.80, tibiae 0.79-0.91, tarsi 1.40-1.97. Tibiae II with 2-4, usually 4, pseudosegments; one specimen has 0, 1 pseudosegments in tibiae IV. Femora I-IV lengths (respectively): 2.86-3.59, 5.61-7.11, 3.00-5.20, 4.03-6.00. Tibiae I-IV lengths (respectively): 2.40-3.60, 4.88-6.84, 2.43-3.28, 3.07-4.82. Ovipositor as in Fig. 16; with single slit sensillum per side on second segment. Seminal receptacles as in Fig. 18; lateral basal loops variable in size and shape, but always appear to have two separate tubes joined at ends. Immatures: Later instars very similar to adult females, but dorsal pattern not as distinct and with numerous white spots. Early instars with same form as later instars, but overall body color yellow brown with only faint brown bands on palpi and legs, no dorsal pattern. Cheliceral tooth generally well developed and pointed. Specimens examined. -CANADA, Alberta ; Can more (Casa’de Mountain ?), Banff National Park, 3-4 September 1939 (H. Schenkel-Rudin), 1 female, 1 immature (NBM); Jasper National Park, The Bald Hills, Maligne Range (2100-2600 m), July-August 1970 (P. Kuchar), pitfall traps near timberline, 4 females, 4 immatures (AMNH), 4 females, 4 immatures (WAS), 1 female, 1 immature (JCC); Castleguard, Crowsnest Pass (3000 m), August 1972 (R. Harmon), 3 males, 3 females (CNC). British Columbia : Mt. St. Paul, mile 392 Alaska Highway (1370 m), 17 July 1959 (R. E. Leech), 1 female, 1 immature (CNC). — < Figs. 20.- Distribution of Liopilio in west- ern Canada and Alaska, squares = L. glaber, circles = L. yukon. 314 THE JOURNAL OF ARACHNOLOGY Liopilio yukon, new species Figs. 2-11, 14, 15, 17, 19,20 Liopilio, n. sp.: Cokendolpher 1980:134. Types.-Male holotype from Isobel Pass (1370 m), mile 206 Richardson Highway, Alaska, 18 July 1962 (R. E. Leech), CNC; and 55 paratypes (listed under specimens examined). Etymology. -The specific epithet is a noun in apposition. Comparisons —Liopilio yukon is similar to L. glaber, but differs by being slightly larger and by having the dorsal abdominal pattern distinct, dorsal pattern obscured or entirely absent in L. glaber. In addition, L. yukon has the last abdominal sternite noticeably indented posteriorly, whereas in L. glaber this sternite is straight or only slightly indent- ed. The male palpal patellae and tibiae of L. yukon are generally armed ventrally with small denticles (Fig. 7), bare in L. glaber. The second segment of the ovipositor has one slit sensillum per side in L. glaber , whereas in L. yukon there are generally two. Distribution.— Alaska in the United States and Yukon Territory in Canada (Fig. 20). Description.— Males: Total length 4.21-5.00, greatest width 2.44-2.83, maximum height 1.87-2.97; cephalothorax creamy yellow with velvety dark brown splotches; abdo- men creamy yellow to bright yellow with distinct dorsal pattern of velvety dark brown (Fig. 2). The yellow pigments of the dorsum appear to fade in alcohol to an off white to very light yellow. Ocular tubercle length 0.37-0.40, width 0.41-0.43, height 0.18-0.21, distance from anterior edge of carapace 0.40-0.50; dark brown with black rings around eyes. Genital operculum light brown. Abdominal sternites creamy white, junctions with rows of few brown to black spots. Chelicerae light brown with darker brown splotches laterally, teeth black, basal segments usually with a small ridge or tooth ventrally (Fig. 4). Palpi (Figs. 6-8) light brown with dark brown stripes on dorsa of femora, patellae, and proximal 2/3 tibiae; dark brown spots in two rows on lateral margins of patellae and tibiae; distal tip of tarsi dark brown; ventral surface of tarsi, tibiae and distal portions of patellae with many dark brown to black denticles, those on tarsi in two rows. Palpal lengths: femora 1.02-1.20, patellae 0.68-0.72, tibiae 0.78-0.84, tarsi 1 .61-1 .78. Legs light brown with dark brown bands preceded by light areas on distal ends of femora, patellae, and sometimes tibiae; tibiae II with 2-5, generally 4, pseudosegments. Femora I-IV lengths (respectively): 3.08-3.40, 6.00- 6.40, 3.20-3.85, 4.73-5.12. Tibiae I-IV lengths (respectively): 2.80-3.20, 5.47- 6.00, 2.94-3.00, 3.41-3.78. Penis as in Figs. 14, 15. Females: Form and coloration essentially as in males, except coxae sometimes with distal band of dark brown and tibiae with median brown band in addition to subdistal band; abdomen white with brown markings. The yellow pigments of cephalothorax appear to fade in alcohol, but a distinct dorsal pattern, brown markings on creamy white, is visible on a specimen over 100 years old. Basal segments of chelicerae with rounded lobes or smooth ventrally (Fig. 6). Total length 5.90-6.83, greatest width 2.64-5.00, maximum height 2.85-4.07. Ocular tubercle length 0.38-0.40, width 0.40-0.45, height 0.20-0.22, distance from anterior edge of carapace 0.34-0.43. Palpi as in Figs. 9-1 1 . Palpal lengths: femora 0.99-1.18, patellae 0.63-0.78, tibiae 0.80-0.88, tarsi 1.64-1.73. Tibiae II with 2-5, generally 3, pseudosegments. Femora I-IV lengths (respectively): 2.40-3.50, 4.23-6.90, 2.37-3.79, 3.70-5.42. Tibiae I-IV lengths (respectively): 2.34-3.13, 4.03-6.02, 2.19-2.91, 2.38-4.00. Ovipositor as in Fig. 17, generally with two slit sensilla per side on second segment, single specimen from Port Clarence with single slit sensillum per side. COKENDOLPHER-THE HARVESTMAN GENUS LIOPILIO 315 Seminal receptacles as in Fig. 19; lateral basal loops variable in size and shape, but always appear to consist of single tube. Immatures: Later instars very similar to adult females, except dorsum gray to gray brown with dark brown pattern; abdominal pattern with several white to gray splotches. Early instars as later instars, except leg bands are indistinct, often only with splotches of brown. Cheliceral tooth pointed (Fig. 3). Specimens examined. -Male holotype and 55 paratypes. UNITED STATES, Alaska: Kotzebue (= Katzebue), 9 September 1959 (A. B. Krom), 1 male, 1 female, 32 immatures (AMNH), Port Clarence, Grantley Harbour (Iman suk, 65°15'N, 166°20'W), 23-26 July 1879 (no. 105 1. Vega Expedition), 1 female, 2 immatures (NHR); Isobel Pass, mile 206 Richardson Highway (1370 m), 17 July 1962 (R. E. Leech), 2 immatures (CNC), 13 July 1962 (P. J. Skitsko), 1 male (JCC), 18 July 1962 (R. E. Leech), 2 males, 1 female (CNC); Dadina River drainage, sec. 28, SW quar. (62°01'N, 144°30'W - 1150 m), alder thicket, 22 July 1978 (R. Saltmarch), 1 male (AMNH). CANADA, Yukon Territory: Kluane National Park, Grizzly Creek (61°05'N, 139°06'W - 1890 m), 27 July 1976 (D. Cossette), 1 female (CNC); North Fork Pass, Ogilvie Mountains (62°21'N, 138°15'W - 1060 m), 18 June 1962 (R. E. Leech), 1 immature (CNC); Swede Dome, 55 km W Dawson (1150 m), 5 June 1962 (R. E. Leech), 5 immatures (CNC), (1150 m), 5 June 1962 (P. J. Skitsko), 4 immatures (CNC), (1190 m), 2 June 1962 (R. E. Leech), 2 immatures (CNC); mile 11 Canol Road (1370 m), 22 March 1962 (R. E. Leech), 1 immature (JCC). ACKNOWLEDGMENTS I would like to thank Dr. O. F. Francke and Mr. W. D. Sissom, Texas Tech University, for their reviews of the manuscript. Thanks are also extended to the following individuals and their museums for loans and gifts of specimens: Dr. C. D. Dondale, Canadian Nation- al Collection of Insects and Arachnids, Ottawa, Ontario (CNC); Dr. T. Kronestedt, Natur- historiska Riksmuseet, Stockholm, Sweden (NHR); Dr. N. I. Platnick, American Museum of Natural History, New York (AMNH); Dr. W. A. Shear, Hampden-Sydney College, Hampden-Sydney, Virginia (WAS); Dr. E. Sutter, Naturhistorisches Museum Basel, Basel, Switzerland (NMB). Dr. R. G. Holmberg kindly sorted unidentified specimens from the Canadian National Collection of Insects and Arachnids. This study was supported in part by The Museum, Texas Tech University. LITERATURE CITED Cokendolpher, J. C. 1980. Comments on Opiliones described from western North American by Schenkel. Ent. News, 91:133-135. Cokendolpher, J. C. 1981. Revision of the genus Trachyrhinus Weed (Opiliones, Phalangioidea). J. Arachnol., 9: 1-18. Forcart, L. 1961. Katalog der Typusexemplare in der Arachnida-Sammlung des Naturhistorischen Museums zu Basel: Scorpionidea, Pseudoscorpionidea, Solifuga, Opilionidea und Araneida. Ver- handl. Naturf. Ges. Basel, 72:47-87. Holm, A. 1970. Notes on spiders collected by the “Vega Expedition” 1878-1880. Ent. Scand., 1:188-208. Rambla, M. 1977. Opilions (Arachnida) de les cavitats de Sant Lloren? del Munt-Serra de l’Obac. Comun. VI Simposium d’Espeleologia, Terrassa, 9-16. Roewer, C. F. 1923. Die Weberknechte der Erde, Systematische Bearbeitung der bisher bekannten Opiliones. Gustav Fischer, Jena, 1116 p. Schenkel, E. 1951. Spinnentiere aus dem westlichen Nordamerika, gesammelt von Dr. Hans Schenkel- Rudin. Verhandl. Naturf. Ges. Basel, 62:28-62. 316 THE JOURNAL OF ARACHNOLOGY Silhavy, V. 1960. Die Grundsatze der modernen Weberknechttaxonomie und Revision des bisherigen Systems der Opilioniden. Verh. XI. Internat. Kongr. Ent., 1:262-267. Star^ga, W. 1978. Katalog der Weberknechte (Opiliones) der Sowjet-Union. Fragm. Faun. Warszawa, 23(10):197-234. Manuscript received July 1980. Coyle, F. A. and W. A. Shear. 1981. Observations on the natural history of Sphodros abboti and Sphodros rufipes (Araneae, Atypidae), with evidence for a contact sex pheromone. J. Arachnol., 9:317-326. OBSERVATIONS ON THE NATURAL HISTORY OF SPHODROS ABBOTI AND SPHODROS RUFIPES (ARANEAE, ATYPIDAE), WITH EVIDENCE FOR A CONTACT SEX PHEROMONE Frederick A. Coyle Department of Biology Western Carolina University Cullowhee, North Carolina 28723 William A. Shear Department of Biology Hampden-Sydney College Hampden-Sydney, Virginia 23943 ABSTRACT The three populations of S. abboti studied exhibit much higher densities than the two of S. rufipes. In S. abboti mating occurs in July, egg laying in August, hatching in September, and the fully equipped and self-sufficient third instar spiderlings develop by November. The tubes of S. abboti are proportionately longer and have a greater aerial length to underground length ratio than do those of S. rufipes. Other differences in tube architecture exist. Both species capture and feed upon a wide variety of ground arthropods and discard prey remains through the upper end of the tube. S. abboti males search for mates in daylight and appear to be generalized wasp-ant mimics. Our observations of S. abboti male behavior suggests that adult female tubes are marked by a contact sex pheromone. INTRODUCTION Despite their covert behavior and patchy abundance, purse-web spiders (Atypidae) have attracted considerable attention from araneologists for at least two reasons. First, atypids possess an interesting mixture of primitive and specialized characteristics, their outstanding specializations being part of a unique strategy of capturing prey through the walls of a silk tube, the purse-web. Secondly, these spiders are found in north temperate zone countries where other mygalomorph taxa are poorly represented but where arane- ologists are relatively abundant. Major contributions toward understanding the natural history of European Atypus species have come from Enock (1885, 1892), Bristowe (1933, 1958), Ehlers (1937), Gerhardt (1929, 1933), Clark (1969), and Kraus and Baur 318 THE JOURNAL OF ARACHNOLOGY (1974). Knowledge of North American Sphodros species is based primarily upon the observations of McCook (1888), Poteat (1890), Gertsch (1936, 1949, 1979), Muma and Muma (1945), Bishop (1950), and Gertsch and Platnick (1980). In order to learn more about Sphodros behavior and ecology, we spent seven days in 1973 (July 15 to 19 and October 5 and 6) observing Sphodros abboti Walckenaer and Sphodros rufipes (Latrielle) populations at five localities. S. abboti was studied at Mill- pond Plantation on the edge of Thomasville, Thomas Co., Georgia; at Suwannee River State Park, Suwannee Co., Florida, and at Mud Springs, near the Welaka Reserve of the University of Florida, Putnam Co., Florida. S. rufipes was observed at Florida Caverns State Park, Jackson Co., Florida, and at Torreya State Park, Liberty Co., Florida. In this paper we present and discuss these observations if they extend or clarify earlier observa- tions, if they affect existing hypotheses, or if they break new ground. SPHODROS ABBOTI Habitat, microhabitat, and population density. -All three S. abboti populations were in the magnolia forest community of Shelford (1963). At Thomasville and Suwannee River the herbaceous ground cover was dominated by poison ivy (Rhus), Virginia creeper (Parthenocissus), ferns, and greenbrier (Simlax), whereas in the hammock community at Mud Springs, there was a dense cover of saw palmetto ( Serenoa ). At all localities the soil was a soft, grey, very sandy loam with much humus. The soil was humid but well drained, and was covered by a continuous layer of leaf litter. S. abboti tubes were absent from low stream bank areas subject to periodic flooding. Population densities were high at all three localities, being on the order of several hundred tubes per hectare. Commonly, an occupied tree had more than one tube attach- ed to it, and occasionally five or more tubes were attached to a single tree with only 0.5 to 4 cm separating some tubes (Fig. 1). At Thomasville, one 15-cm diameter post oak held one adult female tube and six other tubes nearly as large. Several larger trees held between twelve and fifteen tubes apiece. We observed, as did McCook (1888), that both adult and immature tubes were attached to a variety of hardwood tree species of all sizes, including saplings less than 4 cm in diameter. Because only a few of the pine trees examined supported tubes (pine trees were fairly common at Thomasville), it appears that hardwood tree trunks are more suitable microhabitats than pine trunks. Since the tubes are permanent constructions, added to over the life of the spider, it is likely that those spiders which attach tubes to pines might not be able to maintain them because of the tendency of the bark of most pines in the area to break off in large flakes, or plates. Phenology and spiderling development.— Our discovery at Thomasville on July 16 of two S. abboti males searching for females indicates, in accordance with the observations of Gertsch (1936) and Bishop (1950), that males emerge from their tubes and search for mates in July. Evidently, egg sac construction and egg laying occur in August. This is indicated by the absence of egg sacs and spiderlings in all adult female tubes (N = 15) collected from Thomasville and Suwannee River on July 15, 16, and 19 and by the presence of second instar spiderlings in most of the adult female tubes collected at Suwannee River and Mud Springs on October 5 and 6. An examination of six broods collected reveals much about the timing and pattern of postembryonic development in S . abboti. Five of the six broods consisted of second instar spiderlings. In the remaining brood of 89 from Suwannee River, 26 spiderlings were COYLE AND SHEAR-OBSERVATIONS ON TWO SPHODROS 319 Figs. 1-4. -Tubes of purse-web spiders: 1, four tubes of Sphodros ahboti on a large Magnolia trunk near Thomasville, Ga.; 2, upper end of a tube of S. abboti: 3, tube of Sphodros rufipes on an unidentified hardwood tree at Florida Caverns State Park, Marianna, Fla.; 4, upper end of a tube of S. rufipes. 320 THE JOURNAL OF ARACHNOLOGY very late in the second instar (third instar setae were visible beneath the outer cuticle), two were shedding the second instar cuticle, and 61 were in the third instar. One of the second instar broods from Suwannee River was kept alive at room temperature for 21 days and by the end of that period had developed to the third instar. These spiderlings were more heavily pigmented than the third instar spiderlings collected on October 6. The second instar (= first postembryonic stage of Yoshikura 1958) spiderlings are quite motile and possess weil developed chelicerae and fangs, tarsal claw teeth, and numerous long setae, but they lack spines on the endities and have approximately ten spinneret spigots. Third instar spiderlings possess a more elongate body, many spines on the inner margin of the endites, and approximately 45 spinneret spigots. Yoshikura (1958) reported that the first instar (= Yoshikura’s deutovum) of Atypus karschi (Donitz), which emerges from the embryonic cuticle and chorion at hatching and molts ten days later to the second instar, possesses stubby non-functional chelicerae and fangs, short tarsal claws without teeth, very short and widely scattered setae, and no spinneret spigots. These data indicate that hatching in these populations of S. abboti occurs in Septem- ber. Probably by late September the first instar spiderlings molt to second instar spider- lings, which then emerge from the egg sac. No egg sac remnants were found in any of the tubes containing spiderlings. The molt to the third instar takes place during October. These third instar spiderlings are anatomically equipped to handle prey and construct tubes, and, when placed on soil, they construct small silk tubes resembling the tubes of adults. Abbot’s finding of a large number of spiderlings in their mother’s burrow in November (McCook 1888) indicates that they overwinter there before dispersing in spring, but other authors (McCook 1888, Gertsch 1936, 1979) suggest that dispersal may occur in their first autumn. Brood size.— The mean number of spiderlings in the six broods collected at Suwannee River and Mud Springs was 79.7 (range = 49-142, s. dev. = 35.98). Tube architecture.— The tubes of 24 adult female S. abboti and of a smaller number of immature specimens were excavated, measured, and closely examined in the laboratory. As shown in Figs. 1, 2, and 5, these silk tubes consist of an underground portion which lines the burrow and an aerial portion which extends approximately straight up the tree trunk. The aerial portion of the tube is firmly attached to the bark and held taut by numerous silk attachment strands which are concentrated at its upper end (the upper 3 to 5 cm of the tube) and which often form broad apical attachment bands. Attachment strands are rare or absent below the upper one-third of the aerial part of the tube. This mode of tube attachment probably dampens prey-generated vibrations less than if the entire length of the tube were anchored to the trunk. Contrary to the statements of Abbot (McCook 1888), McCook (1888), and Gertsch (1949) that the underground portion of the tube of S. abboti is as long or longer than the aerial portion, we found (Table 1) that the underground portion of the adult tube is usually well under one-half the length of the aerial portion. The aerial tube length/ underground tube length averaged highest in the Thomasville population (mean = 3.4) and lowest in the Suwannee River population (mean = 2.3); the Mud Springs population was intermediate (mean = 2.7). The burrow, which is approximately circular in cross section, extends straight down into the ground, unless, as is often the case, roots cause it to turn towards the horizontal. Frequently the tube is enlarged just below the soil surface. Possible functions for this chamber are to house the egg sac (as in Atypus af finis, Eichwald [Enock 1885, Bristowe 1958]), to serve as a mating chamber, to allow more efficient prey handling during feeding, and to allow the spider to turn around more easily. COYLE AND SHEAR-OBSERVATIONS ON TWO SPHODROS 321 The bottom end of the burrow tapers and is occasionally two-branched or has one or two collapsed, apparently abandoned, branches. Usually the aerial portion of the tube is nearly cylindrical close to the ground but becomes more flattened as it ascends the tree. A few tubes are nearly cylindrical through- out most of their length; others are more flattened than usual and have a collapsed look over much of their length. Because the tube is flatter near the upper end, its width (maximum diameter) remains approximately constant or increases slightly from ground surface to near the attachment zone, even though its circumference decreases slightly over the same distance. In the attachment zone the tube usually narrows considerably. A few tubes have collapsed side branches attached to the aerial portion. These apparently are tubes which were disloged and have been replaced. The entire inner surface of the tube is smooth white silk. The silk is thicker and the tube is consequently tougher underground than it is aboveground. Soil particles (mostly bits of humus) are enbedded in the outer surface of the aerial portion of the tube. Frequently there is a partial covering of moss. As a result, the color of the emergent tube is primarily light to medium brown or greenish brown. Silk and soil particle density decrease with distance from the ground surface, so that the tube is darkest at its base and becomes lighter distally. In the attachment zone there is usually such a low density of soil particles in the tube that it is very light brown or white. Attachment strands are white. In Comstock (1940) the claim is made that the tube is camouflaged by virtue of matching the bark to which it is attached and that this camouflage is the result of “minute bits of bark, lichens, and moss, which are evidently collected by the spider from the trunk of the tree and fastened to the surface of the web.” Our observations of tube materials and tube construction behavior show that non-green materials in the tube are mainly soil particles excavated from the burrow and implanted into the tube from inside the tube (See also McCook 1888, and Gertsch 1979). Since the sand grains do not adhere to the silk as well as the lighter, flatter, and more irregular humus particles, the latter predominate. The form and distribution of the moss on the tubes indicate that it is not transplanted by the spider but instead reaches the tube surface as spores. The moss grows in a thin layer on the tube surface and is not embedded in the silk. It usually decreases more rapidly in abundance with increasing distance from the base of the tube than does soil particle density. Only the protonema stage (often with buds) of the moss life cycle occurs on the upper (newer) part of the aerial portion of the tube. Actually, by human visual standards S. abboti tubes do not always match the adjacent trunk surface very closely in color. Because of this and their form, they look more like sticks or vines than a part of the bark surface. It appears likely to us that, in addition to reducing visual and tactile identification of the tube by potential predators and prey, the non-silk materials in the tube may also help to stiffen the tube wall and consequently help keep the tube expanded so that prey generated vibrations are not dampened and so that the spider’s approach does not cause gross changes in the tube shape that could warn its prey. Trash Disposal.— Prey exoskeletons and S . abboti molts are commonly found hanging loosely from the outside surface of the upper end of the tube. Since trash is rarely embedded in the tube silk and since trash is sometimes found on the ground near the base of the tube, it appears that the spider does not normally actively attach trash to the tube but that it simply lodges there as it is being pushed through the end of the tube. The thin exoskeletons of prey like spiders and crickets are finely fragmented and these fragments 322 THE JOURNAL OF ARACHNOLOGY are bound tightly together by silk into a compact ball. The harder exoskeletons of other prey (millipeds, beetles, isopods, etc.) are not so finely fragmented and these large pieces of exoskeleton are more loosely bound with silk. Molts are not fragmented and are bound with only a small amount of silk. No trash was found in the bottom end of any tubes. Prey.— The following prey were identified from the trash attached near the tops of S. abboti tubes: 6 spiders (1 Clubionidae), 11 isopods, 6 millipeds (3 Julida, 3 Poly- desmidae), 1 cricket, 11 beetles (3 Staphylinidae, 2 Carabidae, 1 Lampyridae, 1 Cur- culionidae), 7 worker ants, 1 wasp ( Vespula ) and 5 unidentifiable insects. One slug caterpillar (Limacodidae) had been killed but was discarded undigested. Evidently the primary prey of S. abboti, in the fashion of trapdoor spiders, are arthropods which frequent the ground surface, but unlike most trapdoor spiders, S. abboti captures some diurnal aerial insects like Vespula. Reproductive Behavior.— Although courtship and mating behavior have been observed in some European Atypus (Enock 1885, 1892, Gerhardt 1929, 1933, Ehlers 1937, Bris- towe 1958, Clarck 1969), there is virtually no evidence for how males locate and recognize adult female tubes. There is no published data on reproductive behavior in American Sphodros species. Consequently, we were fortunate to observe two S. abboti males late in the afternoon of July 16 near Thomasville. The first male was discovered inside the upper end of an adult female tube. As the tube was excavated, this male escaped through a slit in the top of the tube. The second male was spotted as it walked over the leaf litter near the base of the tree from which an adult female tube (the tube in which the above male was found) had just been removed. In behavior (rapid jerky movements), color (black with an iridescent purple abdomen), and form (relatively slender body and legs) the male resembled a large, black ant or a pompilid wasp. When it reached the tree trunk, it walked over its surface, and when it arrived at the position where the tube had been minutes before, it immediately turned Figs. 5 -6. -Diagrammatic composite drawings showing the major features of the tube of purse-web spiders: 5, Sphodros abboti ; 6, Sphodros rufipes. a, zone of attachment; b, soil level. See text for further explanation. COYLE AND SHEAR-OBSERVATIONS ON TWO SPHODROS 323 and walked rapidly up the trunk in a straight path coincident with the natural position of the tube. It stopped abruptly where the upper end of that tube had been attached. It then pivoted back and forth as if searching for the tube. Then it walked down and laterally away from this point and stopped briefly on a medium sized tube, cut a slit in its upper end, but did not enter. It then moved around the trunk to another medium sized tube and examined it briefly before leaving it. This male was then placed on the trunk of another tree about 15 cm from the undisturbed tube of an adult female. It walked around the trunk, stopped when it encountered this tube, turned, and walked up the tube to its top. It quickly cut a slit in the top with its fangs and immediately entered the tube. After hesitating in the top, it then descended rapidly. As the tube was excavated, the male returned to the top and escaped through the slit. This male was later placed on yet another trunk near the spot where another adult female tube had been removed. It walked around the trunk, turned when it reached the position where the tube had been, walked up a path coincident with the tube’s natural position, and stopped where its upper end had been. It then remained motionless for several minutes, was retrieved, and the observations were ended. The evidence suggest that the diurnally wandering males of S. abboti have evolved a defensive mechanism of generalized wasp-ant mimicry in response to selection pressure from visual vertebrate predators. We speculate that diurnal mate searching, which does not seem to be the rule in burrowing mygalomorphs (Main 1957, Buchli 1962, Coyle 1971) and which is not the practice in some Atypus species (Bristowe [1958] calls the male of Atypus affinis “a night wanderer.”), may increase the efficiency of tube finding for males of S. abboti by allowing visual orientation toward tree trunks, which can probably be perceived without large eyes. The nocturnal wandering of A. affinis males (Enock 1885, Bristowe 1958) is consistent with our hypothesis, since visual orientation toward A. affinis tubes, which are not attached to trees, would probably require a very sophisticated visual system. The behavior we observed strongly suggests that adult female tubes of S. abboti are marked by a contact sex pheromone, some of which remains on the bark when a tube is removed. This is the first strong evidence supporting the hypothesis, hinted at by Bristowe (1958) but more completely developed by Platnick (1971), that the prime releaser of courtship and mating in Sphodros is a contact chemical produced by the adult female. It is to be expected that immature spiders do not produce the pheromone. Then why did the male slit open one immature tube? There is at least one interpretation of this behavior which is consistent with the hypothesis that immature tubes lack the phero- mone. Perhaps two stimuli, the sex pheromone and contact with a tube, are required to release the tube slitting response, so that contact with the immature tube, just a few seconds after stimulation by the pheromone of the removed adult female tube, completed the stimulus requirements for tube slitting. Enock (1885) observed that A. affinis males coming in contact with the tubes of mated females “immediately run away as fast as possible,” without receiving an overt signal from the female. Bristowe (1958), however, felt that the ability of the A. affinis male to recognize the reproductive status of the female was based upon the presence or absence of an overt response by her to his drumming on the tube with palpi and legs, a male pattern which we did not see. Bristowe says that “if the female is immature or pregnant his advances from the outside of the tube are repulsed by another signal— a sharp tug or series of tugs at the tube” Obviously there is a need for careful study of Sphodros and Atypus courtship behavior. 324 THE JOURNAL OF ARACHNOLOGY Table l.-Tube dimensions for adult female Sphodros spiders. Range S', abboti (N=24) Mean Std. dev. S. rufipes (N=4) Range Mean Total length (cm) 24-45 31.1 4.82 33-52 41.5 Length of aerial portion (cm) 18-35 22.8 4.11 18-35 25.8 Length of underground portion (cm) 5-13 8.6 1.78 14-17 15.8 Aerial length/ under- ground length 1. 6-5.4 2.8 0.77 1. 2-2.1 1.7 Tube width (max. diam.) 1/3 of its length above ground (mm) 10-20 16.2 1.73 18-24 21.5 SPHODROS RUFIPES Habitat, mocrohabitat, and population density. -At both Florida Caverns and Torreya, S. rufipes is found in mixed hardwood forests (the beech-magnolia forest of Braun [1950]) which share a number of constituents with the mixed hardwood forests of the Southern Appalachians. The Torreya forest is dryer, more open, and has a much sparser ground cover than the forest at Florida Caverns. The soil at both localities is very sandy loam, but is dryer at Torreya. In both populations of S. rufipes tubes were found on a variety of hardwood trees species, but there appeared to be a decided preference for small trees, a relationship which Muma (1944) also observed. At both localities, in spite of the presence of many larger trees, tubes were found only on trees with diameters from 2-45 cm (mean = 10.4 cm s. dev. = 5.17, n = 10) with all but three of these trees being under 8 cm in diameter. However, as Gertsch (1979) and Jackson et al. (1978) have observed, S. rufipes tubes are sometimes attached to large trees. At Florida Caverns and at Torreya, the S. rufipes tubes were widely scattered, with population densities of less than 50 tubes per acre. Only two trees were found supporting more than a single tube. Each of these trees supported two tubes. Similar densities were reported for the Maryland populatons of S. rufipes studied by Muma and Muma (1945). Thus the demography of S. rufipes appears to be very different from that of S. ahboti. Tube Architecture.— The following description is based upon close examination of excavated tubes of four adult female and six immature S. rufipes specimens. Adult females of S. rufipes , which are much larger than S. abboti adult females, construct wider tubes with a proportionately shorter aerial portion (Figs. 3, 4, and 6; Table 1) The numerous attachment strands of the attachment zone (the upper 3-6 cm of the tube) flare out and form a white sheet often a bit wider than the tube. Below the attachment zone are no more than three attachment points. The tube is flattened near the top and be- comes more rounded as it descends to the ground surface. The width (maximum diame- ter) of the aerial portion is nearly constant throughout its length. Muma (1944) and Muma and Muma (1945) report markedly longer underground tubes for Maryland S . rufipes than we observed in Florida. Perhaps this difference results from harsher soil environments in Maryland. Underground the S. rufipes tube bends slightly to strongly COYLE AND SHEAR-OBSERVATIONS ON TWO SPHODROS 325 toward the horizontal. It is slightly constricted immediately below the soil surface but is enlarged near the bottom end to harbor the egg sac (Muma and Muma 1945) or to serve other functions. The silk is thickest at and below the soil surface, and becomes thinner and more fragile as the tube ascends the tree. Below the mostly white attachment zone, the aerial tube varies from light grey-brown or greenish grey-brown to darker brown or greenish brown, the upper part being lighter than the lower part. As in S. abboti tubes, most of the non-silk material consists of excavated organic soil particles which, because of their greater surface area and lighter density, probably adhere better to the silk than do the sand grains. Two tubes were poorly camouflaged with only a sparse covering of soil material in the upper one-half of their aerial portions. Usually moss is also present on the tube surface, but its form and distribution, as on S. abboti tubes, indicate that it is not transplanted by the spider. Old shriveled aerial tubes lay in the litter attached at the ground surface to two of the adult female tubes. Trash Disposal.— As in S. abboti , large amounts of prey rejectamenta and parts of two molts were found loosely attached to the upper end of S. rufipes tubes. The exoskeletons of soft-bodied prey are macerated into tiny fragments which are then bound together with silk into a tightly packed ball, whereas hard prey exoskeletons are broken into fewer, larger, more loosely bound fragments. The only trash found in the bottom ends of tubes were parts of one molt and pieces of a male S. rufipes exoskeleton. Prey.— The following prey items were identified in the trash collected from S. rufipes tubes: 4 spiders (3 S. rufipes males), 2 isopods, 6 millipeds, 2 crickets, 17 beetles (4 Cerambycidae, 1 Carabidae, 1 Scarabaeidae), 1 ichneumonid wasp, 4 worker ants (1 Camponotus ), 1 caterpillar, and 5 unidentifiable insects. It appears from this and from the data of Muma and Muma (1945) that the diet of S. rufipes , like that of S. abboti , consists mainly of a great variety of ground surface arthropods. Reproductive Behavior.-Collection data accompanying four S. rufipes male specimens we have observed from North Carolina, Tennessee, and Maryland indicate that they search for mates in June and are diurnal, like S. abboti. It would be interesting to witness the behavior of the brightly colored S. rufipes males (red legs; black body, chelicerae, and pedi palps) to see whether mimicry or warning coloration might be involved. The macer- ated exoskeletons of three S. rufipes males found with the tubes of the two adult S. rufipes females indicate that they sometimes fall prey to their (intended?) mates. ACKNOWLEDGMENTS This study has been supported in part by a National Science Foundation grant (GB-34128) to F. A. Coyle, and by a grant to W. A. Shear from the Faculty Research Committee of Hampden-Sydney College. We are grateful for the guidance and hospitality provided by Terry Sedgwick at Millpond Plantation. Dr. Paul Haberland assisted in trans- lating the German papers. We especially thank Lance Hunt for his help in making black- and-white prints from our color slides. LITERATURE CITED Bishop, S. C. 1950. The purse-web spider Atypus abboti (Walckenaer), with notes on related species (Arachnidae: Atypidae). Ent. News, 61(5): 121-124. Braun, E. L. 1950. Deciduous Forests of Eastern North America. Hafner Pub., New York, 596 pp. 326 THE JOURNAL OF ARACHNOLOGY Bristowe, W. S. 1933. Notes on the biology of spiders. - IX.The British species of Atypus. Ann. Mag. Nat. Hist., ser 10, 11:289-302. Bristowe, W. S. 1958. The World of Spiders. Collins, London, 304 pp. Buchli, H. 1962. Note preliminaire sur l’accouplement des araignees mygalomorphes Nemesia caementaria, Nemesia dubia, et Pachylomerus piceus (Ctenizidae). Vie et Milieu, 13(1): 167-178. Clark, D. J. 1969. Notes on the biology of Atypus affinis Eichwald (Araneae - Atypidae). Bull. Brit. Arachn. Soc., 1 (3) : 36-39. Comstock, J. H. 1940. The Spider Book. Rev. and Ed. by W. J. Gertsch, Cornell Univ. Press, 729 pp. Coyle, F. A. 1971. Systematics and natural history of the mygalomorph spider genus Antrodiaetus and related genera (Araneae: Antrodiaetidae). Bull. Mus. Comp. Zool., 141(6):269-402. Ehlers, M. 1937. Neues iiber Vorkommen und Lebensweise der markischen “Vogelspinne,” Atypus affinis Eichw., und iiber die Unterscheidung der deutschen atypus - Arten. Markische Tierwelt., 2(4):257-276. Enock, F. 1885. The life history of Atypus piceus Suly. Trans. Ent. Soc. London, pp. 389-420. Enock, F. 1892. Additional notes and observations on the life history of Atypus piceus. Trans. Ent. Soc. London, pp. 21-26. Gerhardt, U. 1929. Zur vergleichenden Sexualbiologie primitiver Spinner insbesondere der Tetra- pneumonen. Morph. Oekol. Tiere, 14(3):699-764. Gerhardt, U. 1933. Neue Untersuchunger zur Sexualbiologie der Spinnen, insbesondere an Arten der Mittelmeerlander und der Tropen. Morph. Oekol. Tiere, 27:1-75. 1936. The nearactic Atypidae. Amer. Mus. Mov., 895:1-19. Gertsch, W. 1949. American Spiders. D. Van Nostrand, Princeton, 285 pp. Gertsch, W. 1979. American Spiders. 2nd ed. Van Nostrand Reinhold, New York, 274 pp. Gertsch, W., and N. Platnick. 1980. A revision of the American spiders of the family Atypidae (Araneae, Mygalomorphae). Amer. Mus. Nov. (2704): 1-39. Jackson, J. F., J. A. Pounds, and D. A. Rossman. 1978. Trans-Mississippi River localities of the purse-web spider Atypus bicolor Lucas (Araneae: Atypidae). Proc. Louisiana Acad. Sci., 41:17-18. Kraus, O. and H. Baur. 1974. Die Atypidae der West-Palaarktis: Systematik, Verbreitung, und Biologie (Arach.: Araneae). Abh. Verk. Naturwiss. Var. Hamburg, 17:85-116. McCook, H. C. 1888. Nesting habits of the American purseweb spider. Proc. Acad. Nat. Sci. Philadel- phia, pp. 203-220. Main, B. Y. 1957. Biology of Aganippine trapdoor spiders (Mygalomorphae Ctenizidae). Australian Journ. Zool., 5(4):402-473. Muma, M. H. 1944. A report on Maryland spiders. Amer. Mus. Nov. 1257:1-14. Muma, M. H. and K. E. Muma. 1945. Biological notes on Atypus bicolor Lucas (Arachnida). Ent. News, 56(5):122-126. Platnick, N. 1971. The evolution of courtship behavior in spiders. Bull. Brit. Arach. Soc., 2(3):40-47. Poteat, W. L. 1890. A tube-building spider. Elisha Mitchell Sci. Soc., 6(2): 134-1 47. Shelford, V. E. 1963. The Ecology of North America. Univ. Illinois Press, Urbana, 610 pp. Yoshikura, M. 1958. On the development of a purseweb spider, Atypus karschi Donitz. Kumamoto Jour. Sci., ser. b, 3(2) : 7 3-86. Manuscript received June 1980, revised August 1 980. Aitchison, C. W. 1981. Feeding and growth of Coelotes atropos (Araneae, Agelenidae) at low tempera- tures. J. Arachnol., 9:327-330. FEEDING AND GROWTH OF COELOTES ATROPOS (ARANEAE, AGELENIDAE) AT LOW TEMPERATURES C. W. Aitchison Department of Entomology University of Manitoba Winnipeg, Manitoba R3T-2N2, Canada ABSTRACT Feeding by Coelotes atropos decreases sharply at and below 6°C compared to that at 8°C and 10°C. As temperature decreases, each individual requires less energy in terms of calories. No growth occurred below 8°C, suggesting a developmental zero between 6° and 8°C. INTRODUCTION Most research on feeding and growth of spiders has been conducted at temperatures between 10° and 20°C (Miyashita 1968, Hagstrum 1970, Petersen 1971 , Schaefer 1977, Workman 1978). However, some species which are active under snow in southern Canada may feed and possibly even grow at temperatures close to 0°C (Aitchison 1978). Coelotes atropos (Walckenaer) constructs silk-lined tunnels under stones, with a collar of silk at the main entrance (Bristowe 1958). Adults mate in spring and early summer, then the female deposits her egg cocoon in a tunnel in June, guarding it and later the spiderlings for the rest of her life (Bristowe 1958). This paper describes quantitative feedings and growth of C. atropos collected from the field and reared at constant tem- peratures from 2° to 10°C. The study was undertaken to ascertain the extent of feeding and growth of this spider during winter months in northern England. AREA AND METHODS The animals were collected from an ash-sycamore wood ( Fraxinus excelsior-Acer pseudo plana tus), on the west side of Lake Windemere, Ferry House, Keswick, Cumbria, England. Young spiders, instar III or IV, were collected with an aspirator from tunnels on 22 August 1977, when the mean ground temperature was 13.5°C. All spiders were trans- ported to the laboratory at the University of Manchester in plastic bags containing damp filter paper. At the laboratory the spiders were placed in individual vials, 4 dram size, with 1 cm of damp sand at the bottom and fitted with cork stoppers. The animals were temporarily stored at 12°C in groups of 25 spiders each. The temperature was lowered by 328 THE JOURNAL OF ARACHNOLOGY 2°C every 2 days, one group being retained at each of the test temperatures, 10°, 8°, 6°, 4°, and 2°C. The temperatures in the incubators varied ±1°C. Adults of vestigial-winged Drosophila melanogaster Mg. were fed to the spiders, 2 to each of the juveniles and 4 or 5 to each of the adults every two days (about maximum consumption). The number of flies eaten was determined by the shrivelled remains; these, and any other dead flies, were removed at feeding times. The dates of molts and deaths were noted. Growth was determined by the number of molts per individual which occurred at each temperature. Measurement of the length of the first leg was calculated on each exuvium. RESULTS Several incubator failures necessitated analysing only the data between days 120 and 290, providing 17 complete 10-day periods. There were 21 juveniles at 10°C, 10 at 8°C, 10 at 6°C, 20 at 4°C, and 19 at 2°C. Since there were only 2 or 3 adults at each temperature, the sample size was too small to draw any meaningful conclusions. Thus only the analysis of data on juveniles is considered. The number of flies eaten per 10-day period was analysed using a repeated measures analysis of variance. The analysis revealed significant temperature differences (P<0.01). The least significant difference (LSD) multiple range test revealed significantly more flies were eaten at 8° and 10°C than at 2°, 4°, and 6°C (P < 0.01), averaged over 10-day periods (Figure 1). Furthermore there were no significant differences in consumption rates between the temperatures 2°, 4°, and 6°C or between 8° and 10°C. The mean number of flies eaten averaged per individual per time period was as follows (± the pooled estimate of standard deviation): 2°C, 0.43 ± 1.38 flies; at 4°C, 0.57 ± 1.38; at 6°C, 0.96 ± 1.38; at 8°C, 6.19 ± 1.38; and at 10°C, 5.85 ± 1.38 flies. Analysis of time periods revealed significant differences across time periods (P < 0.01) as well as a significant interaction of time periods and temperature (P < 0.01). There was no apparent trend to the shape of the response across time however. From the mean number of flies eaten, an estimate of the number of calories consumed at each temperature may be made. Assuming that each fly averages a live weight of 1.0 mg fresh weight/fly or the equivalent of 1.07 calories (Edgar 1971a), then at 2°C in a period of 10 days, a juvenile spider consumed a mean of 0.46 ± 1.48 calories; at 4 C, a mean of 0.61 ± 1.48 calories; at 6°C, 1.03 ± 1.48 calories; at 8°C, 6.62 ± 1.48 calories; and at 10°C, 6.26 ± 1.48 calories. The measurements of the first leg over time do not provide adequate data for separa- tion into instars, and only the number of molts per individual at each temperature is considered. At 10°C, 1 1 juveniles molted once, 10 twice and one specimen three and four times, while at 8°C, 14 juveniles molted once, 7 twice and one individual three times. This contrasts sharply with the molts at the lower temperatures; in the test time between 120 and 290 days there were no molts. DISCUSSION The effects of temperature have a direct result on the food consumption and basal metabolic rate in poikilotherms. A high food consumption was demonstrated very clearly by the juveniles of C. atropos held at 10° and 8°C, while those at 6°C and below ate AITCHISON-FEEDING AND GROWTH OF COELOTES 329 significantly less. Subadults of Pardo sa lugubris Walckenaer held for 140 ± 8 days at 4°C, the length of the inactive period in the field in Scotland, fed very little. It was concluded that winter food consumption was negligible (Edgar 1971b). At 4°C the food intake by juvenile C. atropos was 0.57 ± 1.38 flies eaten/10 days. Lack of growth during winter is associated with most invertebrates (Edgar 1971b, Workman 1978). The last molt of Lycosa T-insignita Boes. et Str. was inhibited by low winter temperatures and possibly by hunger until spring (Miyashita 1968). The third overwintering instar of Tarentula kochi Keyserling is active down to 6°C and has a developmental zero of 10°C (Hagstrum 1970). In the time period of 120 to 290 days, only those juveniles of C atropos held at 8° and 10°C molted, while those at lower temperatures had their molts inhibited. The significant differences between the feeding of spiders held at 8°C and above and that of those held at 6°C and below suggest that there is a major physiological change in the metabolism of Coelotes at around 7°C. The significantly different food consumption and lack of growth at lower temperatures also support this hypothesis. Hence it may be surmised that the developmental zero occurs between 8° and 6°C for this species, with locomotory activity feasible down to 2°C. Another interesting point is that the maximum survival of adults occurred at 6°C. There can be considerable differences in the developmental zeros of various species. For example, Schaefer (1977) found that Allomengea scopigera (Grube), which matures in the autumn, grows in a broad temperature range and has a developmental zero of -4°C, while Thanatus striatus C. L. Koch (stenochronous) is dependent upon high temperatures for development and will not develop below about 12°C. In this case C. atropos appears to have an intermediate developmental zero; it is obviously not adapted to low tem- peratures as is A scopigera. DAYS IN CULTURE Fig. l.-The mean number of flies eaten per spider per ten days by Coelotes atropos juvenile at different temperatures. 330 THE JOURNAL OF ARACHNOLOGY During winter months in the field spiders would be at an advantage, having a reduced food consumption when the availability of prey is at its lowest point, and yet still be reasonably mobile. Their mobility and occasional feeding at 2°C would help them survive some of the lower temperatures to which they are exposed, although under stones on the soil surface the microclimate would be several degrees warmer than the ambient air temperature. ACKNOWLEDGEMENTS I wish to thank P. D. Gabbutt for helping collect and feed the spiders, the University of Manchester for the use of incubators and laboratory space, D. Sabourin for statistical analysis, G. W. Uetz (University of Cincinnati) and L. B. Smith (Canada Agriculture Research Station, Winnipeg, Canada) for critical reading of the manuscript. LITERATURE CITED Aitchison, C. W. 1978. Spiders active under snow in southern Canada. Symp. Zool. Soc. London, 42:139-148. Bristowe, W. S. 1958. The World of Spiders. Collins, London. 304 pp. Edgar, W. D. 1971a. Aspects of the ecological energetics of the wolf spider Pardosa ( Lycosa ) lugubris (Walckenaer). Oecologia, 7:136-154. Edgar, W. D. 1971b. Seasonal weight changes, age structure, natality and mortality in the wolf spider Pardosa lugubris Walck. in central Scotland. Oikos, 22:84-92. Hagstrum, D. W. 1970 Ecological energetics of the spider Tarentula kochi (Araneae: Lycosidae). Ann. Entomol. Soc. Amer., 63(5): 1297-1 304. Miyashita, K. 1968. Growth and development of Lycosa T-insignita Boes. et Str. (Araneae, Lycosidae) under different feeding conditions. Appl. Entomol. Zool., 3(2) :8 1-88. Petersen, H. M. 1971. Nogle undersogelser over Lycosa arenicolas vaekst og fodebiologi. Flora og Fauna, 77(2):25-34. Schaefer, M. 1977. Untersuchungen iiber das Wachstum von zwei Spinnenarten (Araneida) im Labor und Freiland. Pedobiologia, 17:189-200. Workman, C. 1978. Individual energy budget of Trochosa terricola Thorell (Araneae: Lycosidae) under constant and fluctuating temperature conditions. Symp. Zool. Soc. London, 42:223-233. Manuscript received July 1980, revised September 1980. Platnick, N. I. 1981. On the spider genus Odontodrassus (Araneae, Gnaphosidae). J. Arachnol., 9:331-334. ON THE SPIDER GENUS ODONTODRASSUS (ARANEAE, GNAPHOSIDAE) Norman I. Platnick Department of Entomology The American Museum of Natural History Central Park West at 79th Street New York, New York 10024 ABSTRACT Zelotes javanus (Kulczynski) from Java and Drassodes ciusi Berland from New Caledonia are transferred to the genus Odontodrassus Jezequel, previously known only from West Africa. The latter species is newly synonymized with the former, which is newly recorded from the Philippine, Solomon, and Marshall Islands, and from Jamaica. Other records greatly increase the known range of Odon- todrassus in continental areas. INTRODUCTION While engaged in a continuing series of revisionary studies of American Gnaphosidae, a few specimens of a peculiar species were first encountered among spiders collected in Jamaica by the late Dr. A. M. Chickering. It was immediately obvious that this species does not belong to any of the genera previously recorded from the New World. Although other gnaphosid genera are known to be endemic to the West Indies (for example, the gnaphosine genus Microsa Platnick and Shadab), the absence of any American group with which the Jamaican species might be considered to be closely related suggested that the species may be introduced. This possibility was greatly enhanced when additional speci- mens of the species were found in a collection of gnaphosids from Eniwetok Atoll in the Marshall Islands made available by Dr. J. A. Beatty. A subsequent survey of available Old World gnaphosid material has revealed (1) that the Jamaican specimens belong to a species which is widespread in the Pacific and which has been described at least twice in genera to which it does not belong, (2) that the species is congeneric with the type species of Odontodrassus , described from West Africa by Jezequel (1965), and (3) that Odontodrassus is an extremely widespread (and easily recognizable) genus that will probably prove to contain numerous species previously described in the various “wastebasket” groups (like Drassodes) which currently contain large numbers of unrelated species. It is hoped that this paper will lead arachnologists to discover Odontodrassus specimens in collections of (and species in the large, if not par- ticularly useful, literature on) Old World gnaphosids. 332 THE JOURNAL OF ARACHNOLOGY Material has been studied from the collections of the American Museum of Natural History (AMNH), Dr. J. A. Beatty (JAB), the British Museum (Natural History) courtesy of Mr. F. R. Wanless (BMNH), the California Academy of Sciences courtesy of Dr. D. H. Kavanaugh (CAS), and the Museum of Comparative Zoology courtesy of Dr. H. W. Levi (MCZ). The illustrations are by Dr. M. U. Shadab of the American Museum. Odontodrassus javanus (Kulczynski), new combination Figs. 1-4 Scotophaeus javanus Kulczynski 1911:470, Figs. 21, 24 (female holotype from Buitenzorg [= Bogor], West Java, Java, may be in Polska Akademia Nauk, not examined). Drassodes ciusi Berland 1924:192, Figs. 54-56 (male holotype from Ciu, New Caledonia, may be in Museum National d’Histoire Naturelle, not examined), 1929:393, Figs. 3, 4; Marples 1960:385. NEW SYNONYMY. Zelotes javanus: Reimoser 1927:1, 1929:1; Mohr 1930:295. Discussion. -Detailed descriptions of this medium sized (length up to 5 mm) species have been provided by Kulczynski and Berland and will not be duplicated here. The previous placements of the species are not defensible. Kulczynski (1911) indicated that his assignment of the species to Scotophaeus was highly questionable, and Berland (1929) admitted that it is unlikely that any true Drassodes occur in the Australian area. Reimoser (1929) justified his transfer of the species to Zelotes by reference to the eye arrangement and cheliceral dentition, but neither the metatarsal preening comb nor the type of gen- italia characteristic of Zelotes are present. Placement of the species in Odontodrassus is supported by the structure of both the male and female genitalia. The type species of Odontodrassus , O. nigritibialis Jezequel (1965:296, Figs. 3, 4a-c), resembles O. javanus in having males with a greatly enlarged embolar base and an extremely elongate, basally originating, and retrolaterally directed Figs. 1-4 .-Odontodrassus javanus (Kulczynski), specimens from Jamaica: 1, left male palp, ventral view; 2, left male palp, retrolateral view; 3, epigynum, ventral view; 4, epigynum, dorsal view. VLATNICK-ODONTODRASSUS 333 embolar tip which fits along the retrolateral edge of a large, flat functional conductor that occupies most of the ventral surface of the palpal bulb (Figs. 1, 2); females of both species have an epigynum with a wide median ridge (separating two lateral atria leading to the copulatory pores) that covers two large and externally visible median longitudinal ducts with transverse anterior extensions and rounded posterior connections to small, medially situated spermathecae (Figs. 3, 4). These characters are diagnostic of the genus, although judging from Jezequel’s illustrations (1965: Figs. 5, 6) it is unlikely that O. bicolor Jezequel (the only other species currently assigned to the genus) actually belongs to Odontodrassus. Specimens of O. javanus can be easily distinguished from O. nig- ritibialis by having only a single retrolateral tibial apophysis in males and only a single anterior epigynal hood in females. No significant differences have been detected among males of this species from Jamai- ca and Eniwetok. Females vary in the degree of coiling of the ducts connecting the spermathecae with the median longitudinal ducts, and hence in the orientation of the spermathecae themselves. This variation occurs among females collected at the same place and time, and sometimes between the right and left sides of an individual. Although no specimens of the species have been available from either type locality (or to confirm the records from islands off the coasts of Sumatra and New Caledonia reported by Reimoser and Berland, respectively), the illustrations published by Kulczynski and Berland are sufficiently detailed to allow placement of their material well within the range of varia- tion shown by available females. There are at least four additional species of Odontodrassus from South Africa (CAS), Algeria (AMNH), Israel (MCZ), and Nepal (CAS). All can be easily distinguished from O. javanus by their epigynal hood, which is curved almost into a semicircle and situated more posteriorly than in that species. It seems likely that these (and other) species of the genus have already been described from various localities; for example, the South African specimens may belong to Drassodes ereptor Purcell (1907:310, Figs. 16, 17), and the Nepalese to Drassodes himalayensis Tikader and Gajbe (1975:274, Figs. 1-5). Distribution.— Pacific islands from Java and the Philippines east to New Caledonia and Niue; Jamaica. Parts of this range are probably due to human introductions. Material Examined. -JAMAICA: St. Andrew. Old Hope Road, Liguanea, 8 October 1957 (A. M. Chickering), 1 female (MCZ); St. Catherine : 1.5 mi. SW Spanishtown, 10 October 1957 (A. M. Chickering), 4 males, 3 females (MCZ, AMNH). MARSHALL ISLANDS: Eniwetok Atoll: Japtan Islet (grass clumps in /poraetf-sedge-grass community), 5 July 1968 (J. W. Berry), 1 male, 3 females (JAB, AMNH). NIUE: no specific locality, in open plantation (B. J. Marples), 1 female (BMNH). SOLOMON ISLANDS: Guadalcanal: Lunga River region (F. E. Samson), 1 female (AMNH). PHILIPPINE ISLANDS: Luzon: no specific locality, June-July 1945 (R. B. Burrows), 1 female (AMNH). LITERATURE CITED Berland, L. 1924. Araignees de la Nouvelle Caledonie et des lies Loyalty .In Sarazin, F., and J. Roux, Nova Caledonia, Zoologie, vol. 3. Berlin, pp. 159-255. Berland, L. 1929. Araignees recueillis par Madame Pruvot aux lies Loyalty. Bull. Soc. Zool. France, 54:388-399. Jezequel, J.-F. 1965. Araignees de la Savane de Singrobo (Cote d’Ivoire), IV. Drassidae. Bull. Mus. Natl. Hist. Nat., ser. 2, 37:294-307. Kulczynski, W. 1911. Symbola ad faunam Aranearum Javae et Sumatrae cognoscendam, II. Sicariidae, Dysderidae, Drassodidae, Zodariidae. Bull. Acad. Cracovie, 1911:451-496. Marples, B. J. 1960. Spiders from some Pacific islands, part IV. The Cook Islands and Niue. Pacific Sci., 14:382-388. Mohr, J. C. van der Mer. 1930. Notes on the fauna of Palau Berhala. Treubia, 12:277-298. 334 THE JOURNAL OF ARACHNOLOGY Purcell, W. F. 1907. New South African spiders of the family Drassidae in the collection of the South African Museum. Ann. Mag. Nat. Hist., ser. 7, 20:297-336. Reimoser, E. 1927. Spinnen aus Pulu Berhala. Misc. Zool. Sumatrana, 21:1-4. Reimoser, E. 1929. Die Spinnenfauna von Pulu Berhala. Misc. Zool. Sumatrana, 38:1-3. Tikader, B. K. and U. A. Gajbe. 1975. New species of Drassodes spiders (Araneae: Gnaphosidae) from India. Oriental Insects, 9:273-281. Manuscript received September 1980, accepted October 1980. The Journal of Arachnology 9:335 RESEARCH NOTES NEW PSEUDOSCORPION SYNONYMIES (PSEUDOSCORPIONIDA, CHERNETIDAE AND CHELIFERIDAE) During the preparation of a new key to the genera of North American pseudoscor- pions, it became apparent that a couple of synonymies existed. In order to clarify the taxonomic picture, the following comments are offered. Acumino cherries Hoff Acumino cherries Hoff, 1949, Illinois Nat. Hist. Surv. Bull 24:476; 1958, Amer. Mus. Novitates 1875:25,47; 1961, Bull. Amer. Mus. Nat. Hist. 122:450. Phoberocheirus Chamberlin, 1949, Amer. Mus. Novitates 1430:6; Hoff, 1958, Amer. Mus. Novitates 1875:25,47. NEW SYNONYMY. The genus was first defined by Hoff on the basis of Hesperochernes crassopalpus Hoff from Arkansas in a paper published in June 1949. Later that same year Chamberlin published the description of Phoberocheirus based upon P. cribellus , new species, from Virginia. At the time, it appeared that the large number of sense spots on the medial side of the chelal hand of the male P. cribellus were unique and distinguishing; that supposi- tion has gone unchallenged to the present. I have examined the types of both H. crassopalpus and P. cribellus , which are in the collections of the Illinois Natural History Survey and the American Museum of Natural History, respectively. Study of the holotype male of H. crassopalpus reveals that sense spots are present on the chela just as in P. cribellus. Likewise, all other aspects of the morphology of the two forms are very similar; these include the shape and trichobothrial chaetotaxy of the palpal chela, the placement of the tactile seta on the tarsus of leg IV, the acuminate nature of setae b and sb of the chelicera, and the general structure of the male genitalia. Unfortunately, the type collection of P. cribellus consists only of the holotype male. However, I have at hand other undoubted representatives of this form (including females) from Blount County, Tennessee, and Jackson County, Florida. In all the females the spermathecae are long, thin tubules with expanded ends, just as in the paratype females of H. crassopalpus (Fig. 1). All the evidence indicates that the two described forms are conspecific. Therefore, Phoberocheirus cribellus Chamberlin must be considered a synonym of Acumino cherries crassopalpus (Hoff), and Phoberocheirus Chamberlin is a synonym of Acumino cherries Hoff. In addition, in regard to the distribution of the species A. crassopalpus , it can be noted that presumed females (unaccompanied by the very distinctive males) have been taken in Cook County, Illinois; Transylvania County, North Carolina; and Rabun County, Georgia. Also, a single male of the species has been reported from Multnomah County, Oregon, by Benedict (1978, Ph.D. Dissertation, Portland State University, Portland, Oregon). The Journal of Arachnology 9:336 Fig. 1 .-Acuminochernes crassopalpus (Hoff): spermathecae of female. Levichelifer Hoff Levichelifer Hoff, 1946, Bull. Chicago Acad. Sci. 7:486; 1950, Amer. Mus. Novitates 1448:15; 1956, Amer. Mus. Novitates 1804:8; 1958, Ameri. Mus. Novitates 1875:34,49. Ocalachelifer Chamberlin, 1949, Amer. Mus. Novitates 1430:17. Hoff, 1958, Amer. Mus. Novitates 1875:35,49; 1964, Amer. Mus. Novitates 2198:17. NEW SYNONYNY. A similar situation exists with respect to the genera Levichelifer Hoff and Ocalachelifer Chamberlin. Hoff described the former in 1946, based upon Idiochelifer fulvopalpus Hoff from Tamaulipas, Mexico. Chamberlin’s description of Ocalachelifer , based upon Ocalachelifer cribratus Chamberlin from Florida, appeared in 1949. Hoff (1956, 1964) commented on the similarities between the two genera but continued to regard them as distinct. I have examined the holotype male and allotype female of Idiochelifer fulvopalpus [AMNH] and other specimens of both sexes from Tamaulipas and Texas and the para- type male of Ocalachelifer cribratus [AMNH] and other specimens of both sexes from Florida. Careful study and comparison reveal that there are no essential differences in the characters formerly used to separate the two genera. These differences are resolved as follows: (i) tergal keels present in both forms, though less heavily sclerotized in the eastern specimens; (ii) spirally coiled tubules in the coxal sacs of males equally developed in both; (iii) a sclerotic rod in the statumen convolutum of males in both eastern and western specimens, although sometimes difficult to observe due to position of adjacent structures; and (iv) no spine on the tarsus of the first leg, although the outer distal angle of the tarsus is sometimes roughened in both eastern and western specimens. Altogether, the specimens are very similar in general characters and in the special features of the male genitalia and coxal sacs and the fact that females bear many setae on the pleural mem- branes. The eastern and western forms may be distinguished as separate species on the basis of size, proportions and degree of sclerotization, but they are certainly congeneric. William B. Muchmore, Department of Biology, University of Rochester, Rochester, New York 14627 Manuscript received August 1980, accepted September 1980. The Journal of Arachnology 9:337 THE IDENTITY OF OLPIUM MINUTUM BANKS (PSEUDOSCORPIONIDA, OLPIIDAE) Olpium minutum was described briefly by Banks in 1908 on the basis of specimens from Austin, Travis County, Texas. Because of the brevity of the description and the fact that no further material has been collected, the species has remained in limbo in regard to modern pseudoscorpion taxonomy. Beier (1932) provisionally placed it in Pachy olpium and Hoff (1958) followed that judgement. I have reexamined the types in detail and find that they are referable to the genus Serianus Chamberlin. Serianus minutus (Banks), new combination Figs. 1-3 Olpium minutum Banks, 1908, Bull. Wisconsin Nat. Hist. Soc. n.s. 6:42. Pachyolpium? minutum, Beier, 1932, Das Tierreich 57:196; Hoff, 1958, Amer. Mus. Novitates 1875:16. Material examined.— Three type specimens are in the collection of the Museum of Comparative Zoology, Harvard University. All were mounted on slides. The male (WM 1948.01001) is designated the lectotype; the paratypes are a female and a nymph. They were collected in a nest of “ Eciton coecum ” (= Labidus coecum) at Austin, Travis County, Texas, by C. T. Brues (no date given). Description of adults.— Banks has described the general morphology briefly but accu- rately. In addition, it can be noted that they have the characteristics of the genus Serianus (see Hoff, 1964, Bull, Inst. Jamaica, Sci. Ser. 10 [3] :35). Male and female alike except for genitalia. Most tergites and sternites divided; surfaces of carapace and scuta smooth. All setae delicate and acuminate; carapace with about 17 setae, 4 at both anterior and posterior margins; tergal chaetotaxy 4:4:4:4:4:5:4:4:4:3T2T3:?:2; most sternites with 6 or 7 marginal setae, and in addition, in the male, sternites 6-8 each with a medial group of 4 setae. Chelicera with 5 setae on hand; flagellum of 4 setae; fixed finger with 3 small teeth, movable finger with a typical subapical lobe; galea with 1 large lateral and 2 small terminal rami (Fig. 1); serrula exterior with 18 plates. Palp robust (Fig. 2); femur 2. 7-2.9, tibia 2.0-2.25, and chela (without pedicel) 3.2 times as long as broad; hand (without pedicel) 1. 9-2.0 times as long as deep; movable finger 0.85 as long as hand. Femur with a long tactile seta on dorsal surface in proximal half. Trichobothria as shown in Fig. 3. Fixed finger with about 18 low teeth, only a few with cusps; movable finger with about 16 less well developed teeth. Legs short and robust; leg I with telofemur longer than basifemur, the articulation between the segments virtually immovable. Leg IV with entire femur 2.5 times as long as deep. Arolia longer than tarsal claws and deeply divided. Measurements (mm).— Figures given first for the lectotype male, followed in parenthe- ses by those for the female. Body length 2.02(2.17). Carapace length 0.50(0.535). Chelic- era 0.18(0.185) long. Palpal trochanter 0.225 (0.235) by 0.11(0.125); femur 0.365(0.365) by 0.125(0.135); tibia 0.36(0.35) by 0.16(0.175); chela (without pedicel) 0.64(0.66) by 0.20(0.205); hand (without pedicel) 0.35(0.38) by 0.185(0.19); pedicel The Journal of Arachnology 9:338 about 0.05 long; movable finger 0.29(0.31 5). Leg I: basifemur 0.09(0.09) long; telofemur 0.155(0.155) long. Leg IV: entire femur 0.38(0.385) by 0.155(0.155); tibia 0.26(0.265) by 0.09(0.09). Remarks.— Serianus minutus is very similar in many respects to both S. dolosus Hoff from New Mexico and S. carolinensis Muchmore from the south-eastern United States; from each it varies in some small details. It is possible that only a single species is represented, ranging across the entire southern part of the United States, though at the present time not enough good material is available to decide this. Serianus argentinae, new name Serianus minutus Hoff, 1950, Arthropoda, Buenos Aires 1:233. With the knowledge that Olpium minutum Banks belongs in the genus Serianus, S. minutus Hoff becomes a junior homonym and must be replaced. The name argentinae refers to Argentina, the country in which the species was discovered. William B. Muchmore, Department of Biology, University of Rochester, Rochester, New York 14627. Manuscript received August 1980, accepted September 1980. The Journal of Arachnology 9:339 USE OF SPIDER THREADS AS RESTING PLACES BY TROPICAL INSECTS As spiders move about, they leave behind silken trail lines or safety threads which gradually accumulate until wind or rain breaks them. Some small flies are commonly found hanging on these abandoned spider silk threads in tropical forests (Fig. 1), espe- cially at evening and night. This behavior probably affords them protection from pre- dators, since it both isolates them from small walking animals like ants, and at the same time facilitates rapid escapes (Eberhard, W. G. 1980. Nat. Hist. 89 (1):56-61). In some places the number of insects hanging on threads is very large, and includes many species of several families of flies (principally Cecidomyidae) as well as occasionally wasps and moths (R. Gagne, pers. comm.); this phenomenon is thus of some importance in biology of many small forests insects. The present report is concerned with some aspects of what kind of threads are chosen by these insects as resting places. Observations were made on 7, 9, and 10 January 1978, at Finca La Selva Research Station, Sarapiquf, Costa Rica. On 7 January a count was made of the number of insects found along the edges of a trail through more or less virgin forest (in the reserve near the buildings). Each abandoned spider thread seen within about 1 m of the trail and from 0 to 2 m above the ground was measured (length) and classified with respect to inclination as horizontal (0-20°), inclined (20-70°), or vertical (70-90°), (angles estimated by eye- sight). On 9 and 10 January, observations were made in three small plots in overgrown cacao groves. Thirty-six forked sticks were strung with trail line threads of different species of spiders stuck into the ground in the afternoon, then checked hourly to determine the Fig. 1. -Insects hanging on abandoned spider silk threads in tropical forest. (Approximately size of left hand insect: 3 mm). The Journal of Arachnology 9:340 Table 1.- Distribution of insects with respect to inclination of threads in natural conditions. Horizontal threads are favored over inclined (p<0.025) and vertical (p<0.005) threads. Inclination Percent of threads occupied by insects Total number of insects Total length of thread(m) Number of insect/m Average length of thread(m) Horizontal 50 (n=27 1) 165 34.95 4.72 0.129 Inclined 39 (n=188) 77 22.49 3.42 0.120 Vertical 35 (n= 74) 27 10.38 2.60 0.140 TOTALS 44 (n=533) 269 67.82 number of insects which settled on each kind of thread. The threads were placed on the frames by inducing the spider to fall free on its trail line, and turning the stick so that the thread was wound onto the arms of the fork. On 9 January, the spiders used were mature females oiNephila clavipes (Linnaeus) (9 stakes), Argiope savigni Levi (3 stakes), Micrathena sp. (9 stakes), and a theridiid (prob- ably Acharaeanea or Tidarren ) which was collected on buildings (3 stakes). In addition, thin white cotton thread was used on nine other stakes. The stakes were placed in three plots, varying their inclination so that some threads were horizontal and others inclined or vertical. On 10 January the same distribution was used, but all stakes were horizontal. Nine stakes each with threads of N. clavipes , Micrathena sp., a theridiid, and with cotton threads were used. The statistical test performed on the data was Chi-Square. The distribution of flies found hanging on threads in natural conditions is shown in tables 1 and 2. There were clear tendencies to favor both horizontal and shorter threads. The data have the weakness that occupied threads were more easily noticed, and the “percent occupied” is thus an overestimate. There is no reason to think, however, that the vertical threads were any more visible than other unoccupied threads, and the insects’ tendency to favor horizontal threads is real (Table 1). On the other hand, longer unocc- upied threads might be more easily noticed than shorter ones, suggesting that the observed tendency is an artifact of our searching behavior. Nevertheless, the preference for short threads is very strong: for example, longer threads (> 20 cm, x= 35.6 cm) would be expected, because of their size, to be more than seven times more frequently occupied than the shortest threads, when in fact they were occupied less than twice as often (Table 2). In balance we suspect that there was a real preference for shorter threads. Table 2. -Distribution of insects with respect to the length and inclination of threads in natural conditions. The values are non-random (p<0.001), using the values of threads 0-4.9 cm long as standard and assuming that threads twice as long would be twice as likely to be occupied. Length of Percent of threads occupied Number of insects thread (cm) per occupied thread Horizontal Inclined Vertical H I V 0-4.9 32.56% (n=43) 33.33% (n=50) 31.25% (n=16) 1.00 1.00 1.00 5. 0-9.9 42.21% (n=95) 43.75% (n=64) 44.00% (n=25) 1.26 1.04 1.00 10.0-14.9 55.36% (n=56) 38.89% (n=36) 36.37% (n=ll) 1.19 1.07 1.00 15.0-19.9 78.57% (n=28) 40.74% (n=27) 22.22% (n= 9) 1.23 1.09 1.00 >20 5 3.06% (n=49) 32.26% (n=31) 30.77% (n=13) 1.30 1.10 1.25 The Journal of Arachnology 9:341 Table 3.-Distribution of insects according to the kind of thread and its inclination for threads on forked sticks. January 9 January 10 Total Type of thread H I V Total Meters of thread Av. number per meter (H,I,&V) H Mof thread Av. number per meter Number of insects/m horizontal thread Av. of both days Nephila 3 0 1 4 5.43 0.75 10 5.66 1.76 1.26 Theridiid 1 0 1 2 1.12 1.78 6 5.34 1.12 1.23 Micrathena 5 0 1 6 4.19 1.43 13 6.28 2.10 1.81 Argiope 0 0 0 0 1.26 0 - — — — Cotton thread 1 1 0 2 5.67 0.35 11 5.67 1.93 1.14 Totals 10 1 3 14 17.67 0.79 40 22.95 1.73 The numbers of insects found on the threads laid on forked sticks are shown in Table 3. Again there was a tendency for more flies to rest on horizontal threads, and for this reason all stakes were positioned to hold the threads horizontally on 10 January. Although the total numbers of insects observed during the two nights were different, the numbers per meter of horizontal thread were almost the same (1.63 and 1.73). These are substantially lower than the numbers per meter of horizontal thread in natural con- ditions (Table 1). Table 3 also shows the distribution of insects on the different types of threads used. Although the threads differed in thickness, strength, and consistency, the number of insects found on them are similar; there were no statistically differences. The behavior of insects hanging on spiders’ silk threads has been reported by Robin- son, M. H. and B. Robinson (1976. Entom. Mon. Mag. 112 : 1-3). However, in the present observations, the insects rest on abandoned individual lines, while the insects studied by those authors, rest on threads that form part of an entire spider’s web. It would seem that the situation here described offers more security because the risk of resting close to a potential predator is eliminated; yet, it still retains the advantage of being associated to a very sensitive line in case of an approach by other animals. This is supported by the fact that in almost all cases the lines were occupied by only Qne fly (Table 2). In summary, the insects’ choice of kind of thread seems to be random. However, with respect to thread orientation, horizontal threads are more attractive than vertical and inclined threads; and shorter threads are probably preferred to longer ones. This project was completed as part of the course, “Ecology of Tropical Aracnids”, organized and coordinated by Dr. Carlos E. Valerio, at the Universidad de Costa Rica. We are indebted to Dr. William G. Eberhard for his help in carring out fieldwork with us and in the preparation of this paper. Enrique J. Lahmann and Claudia Ma. Zuniga, Escuela de Biologia, Universidad de Costa Rica, San Jose, Costa Rica. (Present address of EJL: Division of Biology and Living Resources, Rosenstiel School of Marine and Atmospheric Science, 4600 Rickenbacker Causeway, Miami, Florida 33149, USA.) Manuscript received April 1980, revised September 1980. The Journal of Arachnology 9:342 SYNEMOSYNA BICOLOR IS THE FEMALE OF SYNEMOSYNA AMERICANA (ARANEAE, SALTICIDAE) The salticid genus Synemosyna is one of the better known American genera of antlike spiders. There has been considerable confusion concerning species identification within the genus, because of the great differences in color and pattern between local populations of the same species, or even within the same population. Galiano (M. E. 1966. Rev. Mus. Argentina Cien. Nat., EntomoL, Buenos Aires, 1:339-380) has discussed this problem in her revision of the genus. The Peckhams (G. W. and E. G. 1885. Proc. Nat. Hist. Soc. Wisconsin 1885: 23-42) first described S. americana for two males from Guatemala. S. bicolor was described by them (Peckham, G. W. and E. G. 1892. Occ. Pap. Nat. Hist. Soc. Wisconsin 2: 1-83) for two females from Venezuela. Both were originally described as members of the genus Simonella, now considered a synonym of Synemosyna (Galiano 1966). Pickard-Cambridge (F. 0. 1905. Biologia Centrali-Americana, Araneidea 2: 166-312) sent male and female specimens of Synemosyna to the Peckhams for identifica- tion. They determined these Panamanian specimens to be S. americana. Pickard- Cambridge commented that the females were certainly S. bicolor and that the males seemed to belong with the females. However, he did not believe that the males were S. americana , because of color differences and a few structural differences, none of which involved the palpal tarsus. He also mistakenly illustrated the palpus of S. americana as that of the male of his new species, S. decipiens (the male of which is currently un- known). Galiano (1966) stated that a male identified in the Paris Natural History Museum as S. bicolor was in fact, S. americana , and indicated the possibility that both species on further study might be united. I recently examined a collection of antlike salticids from northeastern Colombia col- lected by John Kochalka. Amongst them were several Synemosyna , including a male, female, and two immatures from the same locality on the same date (Colombia, Dept. Magdalena, Sierra Nevada de Santa Marta, Rio Frio, 533 meters elevation, trail, low- medium vegetation, 23 April 1975). The genitalia of the female is that of S. bicolor , the genitalia and the chelicerae of the male are those of S. americana. Both share the same color pattern, although the male is darker with the yellow areas on the prosoma reduced. A female collected on the same site, but on 1 May 1975, has the genitalia of S. bicolor , with the same color pattern as the specimens taken in April, but with the light prosomal areas somewhat expanded. The color pattern is different than that previously described, and is illustrated in figures 1 and 2. The specimen illustrated is the female collected on 23 April. Figs. 1-2.— Synemosyna americana female from Colombia. Dark areas on prosoma are dark brown, pale areas are yellow. The area of the opisthosoma anterior to the constriction is dark brown with a brown scutum dorsally. This scutum is crossed at midpoint by a narrow band of white scales. The constriction is white, and the posterior portion of the opisthosoma is pale gray brown. 1 , right lateral view; 2, dorsal view. Total length 5.4 mm. 2 The Journal of Arachnology 9:343 On the basis of the specimens, and the contiguous range of the species involved, the following synonymy is established, Synemosyna bicolor (Peckham and Peckham) = Synemosyna americana (Peckham and Peckham) NEW SYNONYMY. Complete previous synonymies and diagnostic illustrations will be found in Galiano (1966). The specimens are in the collections of John Kochalka, and I wish to thank him for the opportunity to examine his collection. Bruce Cutler, 1747 Eustis Street, St. Paul, Minnesota 55113. Manuscript received July 1 980. COLD SURVIVAL OF ARGIOPE AURANTIA SPIDERLINGS (ARANEAE, ARANEIDAE) Salt (1961, Ann. Rev. Ent., 6:55-74) describes insects as being either “freezing toler- ant” if they can survive tissue freezing or “freezing susceptible” if they cannot survive such freezing. In the latter group, and in those species of spiders examined to date, cold survival depends entirely on an ability to lower the supercooling point (that temperature below the freezing point of the body fluids at which spontaneous freezing occurs). For these animals the supercooling point represents the low lethal temperature. In specimens from a European population of Nuctena cornuta (Clerk), improved cold hardiness was demonstrated by a reduction in the supercooling point from about -8°C in summer to -23°C in winter (Kirchner and Kestler 1969, J. Insect Physiol. 15:41-53; Kirchner, 1973 In Effects of temperature on ectothermic organism, W. Wieser ed ., Springer-Verlag, New York). In that study, appreciable quantities (2-3% wet weight) of the cryoprotective compound glycerol were found, but glycerol concentration was not directly related to supercooling point. Recent work by Duman (1979, J. Comp. Physiol. 131 : 347-3 52) has shown that immature overwintering crab spiders ( Philodromus sp.) and sac spiders ( Clubiona sp.) accumulated a protein which influenced cold hardiness and that concentrations of this protein were reflected by the magnitude of thermal hysteresis (a difference between freezing points and melting points of the hemolymph). Duman (1979) noted that supercooling points in these two species were lowest when thermal hysteresis was greatest. He interpretated this correlation as an indication that the protein, along with glycerol (3.3 and 4.4% wt./vol. in the species above respectively) were respon- sible for depressed supercooling points. In the present study, egg sacs of Argiope aurantia Lucas were periodically collected from vegetation in open fields and roadsides about 15 km N of Normal, Illinois. On the following morning egg sacs were exposed for 24 hr to a selected low temperature. Some of the sacs collected 1 December 1978 were kept at ambient conditions above snow until March. Supercooling tests were not feasible on individual spiderlings so estimates were made of percent survival of populations within each egg sac following low temperature exposure. After cold exposure, egg sacs were placed at room temperature for 1-2 days, The Journal of Arachnology 9:344 Table 1.- Survival of Argiope aurantia spiderlings following 24 hr exposure to selected low temperature. Collection date N Temperature (°C) Percent survival (range) 3 November 1978 4 -8 100 - - 3 November 1978 5 -13 100 - - 3 November 1978 6 -25 44.7 (20-69) 24 November 1978 5 -25 85.8 (80-92) 12 December 1978 6 -20 100 - 12 December 1978 6 -25 100 - 12 Decmeber 1978 6 -30 18 (0-50) 12 December 1978 5 -34 0 - 16 March 1979 6 -25 100 - 23 March 1979 14 -25 100 - 13 October 1979 16 -10 100 - 13 October 1979 17 -20 0 - dissected and most spiderlings shaken free. Live spiderlings tended to aggregate which helped distinguish them from dead animals. Simple movement constituted survival. A group of 10-20 live spiderlings was weighed to 0.005 mg and an individual spiderling weight determined. Total weight of spiderlings (living and dead), including those adhering to silk was determined and the total number in the sac estimated by dividing the total weight by the individual spiderling weight. Percent survival was calculated by dividing the number of live spiderlings by the total number. Weight differences between freshly killed and live spiderlings were negligible. Sacs containing individuals in the deutovum stage were not examined. The presence of polyhydric alcohols (glycerol, sorbitol, mannitol) was examined in 6 sacs collected 11 December 1978 and acclimated to -20°C for 5 weeks. Approximately 0.1 g of spiderlings was removed from each sac and macerated in 1.0 ml of water. The resulting fluid was centrifuged and 10 [A of the supernatant applied to chromatography paper. In order to examine the possibility that glycerol present in spiderlings might be destroyed in the maceration procedure, spiderlings from a single sac were macerated in 1.0 ml of 0.1% glycerol rather than water. Spiderling samples along with standard solu- tions were applied to paper and chromatograms prepared (Riddle and Pugach 1976, Cryobiology 13:248-253). Polyhydric alcohols were all detectable to 0.1% wt./vol. by this method. Chromatograms indicated the presence of polyhydric alcohols in the stan- dard solutions and in the sample of spiderlings macerated in 0.1% glycerol, but not in the remaining 5 samples. Table 1 clearly indicates a trend of improving survival to -25°C during November 1978. In animals collected 3 November, some temperature in the range of < -13°C to > -25°C was associated with mortality. In sacs taken 12 December a temperature in the range of <-25°C to > -30°C was apparently lethal. In sacs collected in October 1979, the mortality indicated at some temperature over the range of < -10°C to > -20°C is con- sistent with the improving cold survival evident in the fall of 1978. It is entirely possible that some mortality had occurred in. sacs collected in November and December prior to cold exposure. However, the complete survival of individuals in 9 sacs collected 3 Novem- ber (-8 and -13°C groups) strongly suggests but does not prove, that spiderlings were alive in sacs collected in November prior to cold exposure. Complete survival of spiderlings in The Journal of Arachnology 9:345 12 of the 23 sacs collected on 12 December 1978 and in 16 of the 33 sacs taken on 13 October 1979 similarly supports the inference that spiderlings were alive prior to cold exposure. No mortality occurred among spiderlings in 14 egg sacs exposed above snow during the entire winter. This observation, when considered with ambient temperature records is significant in that it suggests a possible natural improvement of cold hardiness after 12 December 1978. This interpretation is supported by the observation that despite 18% survival at -30°C on 12 December, spiderlings in 20 egg sacs (16 and 23 March samples) all had survived an ambient temperature of -29.4°C about a month later on 15 January 1979. Results of the present study lead to two conclusions. First, they indicate that A. aurantia spiderlings, unlike other spiders which have been examined, do not accumulate polyhydric alcohols during overwintering. Second, they suggest that natural acclimatiza- tion to decreasing ambient temperatures or responses to other environmental factors in the fall and possibly in the winter result in improved cold hardiness. H. Tyler Hillman and Mark Cormier provided valuable assistance in this study. Wayne A. Riddle, Department of Biological Sciences, Illinois State University, Normal, Ullinois 61761 Manuscript received July 1 980, revised September 1980. The Journal of Arachnology 9:346 THE JOURNAL OF ARACHNOLOGY Instructions to Authors GENERAL COMMENTS Manuscripts are acceptable in English, French, Portuguese, and Spanish, and must be typed double or triple spaced throughout. Use good bond paper 8.5 by 11 in. in size, but not erasable bond. Leave ample margins, at least 1.5 in. on the left, and do not hyphenate any words at the right margin. To facilitate prompt review by two or more referees, send the Editor the original manuscript and two carbon or good xerox copies, together with copies of the illustrations, for mailing to referees. Follow the recommendations of Council of Biological Editors ( CBE ) Style Manual, unless indicated otherwise below. Do not edit your own manuscript: italics are permitted only to indicate scientific names; only the TITLE, PRIMARY HEADINGS (e. g. INTRODUCTION, etc.), and RUNNING HEAD should be typed in capital letters. Manuscripts longer than about 1500 words (five double-spaced typewritten pages in Elite, six double-spaced typewritten pages in Pica) should be prepared as Feature Articles, shorter papers as Research Notes. Manuscripts and all related correspondence must be sent to Dr. B. J. Kaston, Associate Editor, 5484 Hewlett Drive, San Diego, California 92115, U.S.A. FEATURE ARTICLES Arrange the various parts of feature articles in the following sequence: (1) mailing address, (2) title, (3) by-line, (4) body of text, (5) acknowledgments, (6) literature cited, (7) figure legends, (8) abstract, (9) footnotes, (10) running head, (11) tables with legends, (12) figures. Put only numbers 1, 2 and 3 above on page 1, and start page 2 with number 4 above, numbering all other pages consecutively. Mailing address.- Include the complete address and the telephone number of that author with whom all correspondence with the Editors should be handled. Title.-When using common or scientific names in the title be sure to include in parenthesis the appropriate higher taxa. Include footnote indication if appropriate (e„ g., to acknowledge grant sup- port), but type footnote separately (see number 9 above). Title should be typed in capital letters and may not exceed 48 characters and spaces per line. By-line. -Include name(s) of author(s) as customarily written (less titles) and complete address(es) including zip code or other postal zone designation. Include footnote indication(s) if appropriate (e. g., to indicate change of address), but type footnote separately (see number 9 above). Leave ample space between title and by-line, and four spaces between name(s) and address(es). Body of text. -Use whatever form seems best to accommodate the necessary description of re- search. Be concise. Conform to CBE Style Manual in all abbreviations. Use the metric system on all measurements (the English system is acceptable only when transcribing locality data accompanying museum specimens), and note that abbreviations of metric units of measurement are not punctuated (e.g., mm and km, but ft. and mi.). Citations should be in the form: Bellrose (1950), or Bellrose (1950:33), or (Bellrose 1950). The complete scientific name of a species or genus, including author(s), must be given the first time they are mentioned in the text. Use single-line notation for fractions (e. g., 1/4 and not li(4-12)/3 and not 4-12). 4 3 The following special directions apply to authors of taxonomic papers: (a) . Do not use abbreviations to indicate that a new name or a new combination is being proposed in a primary heading (e. g., A-us x-us, new species, rather than A-us x-us, n. sp. or comparable abbrevia- tions). (b) . Keys must be typed following the third example given in the CBE Style Manual (3rd ed., p. 66), which is as follows: 1. Use Arabic numerals to designate the leading entry of a couplet 2 Do not designate the second entry of a couplet, either by means of numbers, letters, or other marks 3 2. Type numbers flush to left margin, and start entry on third space. The second, and subsequent, lines of one entry must be indented five spaces STOP The Journal of Arachnology 9:347 (c) . Synonymies must follow the abbreviated style shown below: A-us x-us Jones 1930:3, 1935:9; Russell 1945:453; Smith 1954a:16, 1954b:678; Cooper and Lim 1955:18 (in part). A-usy-us Bates 1932:18, fig. 4. NEW SYNONYMY. A-us z-us: Miranda 1948:98 (misidentification); Harris 1951:3 (in part ?). (n.ec A-us z-us Zimmer (d) . Lists of specimens examined of a given taxon must be the last item typed in the treatment of that taxon as they will be set in smaller type. Adhere to the following style for listing specimens examined: Country: state or comparable political subdivision; county or district, detailed locality (elevation), 14 July 1945 (collector), 2 males, 5 females (acronym of institution where specimens are deposited), next detailed locality within that county, and so forth; next county in the same state; and so forth: next state in the same country: and so forth. Next country: and so forth. Punctuation rules are very simple. Use a period to separate countries, colon to separate states, semi-colon to separate counties, and commas to separate specific localities. Acknowledgments.- Avoid overlooking persons who have in some substantial way assisted with the work. Authors of taxonomic papers should spell out the name, and indicate parenthetically the acronym, of institutions where specimens studied are deposited. Literature cited.-Include only those publications to which reference is made in the text. Adhere to the CBE Style Manual or refer to a previous issue of The Journal of Arachnology for style. Do not abbreviate place names in journal citations. Repeat name(s) of author(s) in case of multiple entries. Figure legends.-Provide one legend for each illustration to be reproduced singly, or for each “plate” consisting of several illustrations. Adhere to the following style: Figs. 1-4 -A-us x-us, male from Timbuktu: 1, left leg; 2, right chelicera; 3, dorsal aspect of genitalia; 4, ventral aspect of abdomen. Type all figure legends consecutively on same page(s), using double space within each legend and leaving 4 spaces between legends. Keep in mind that 85 characters and spaces represent one printed line; for each line subtract 4 mm from the maximum length permissible on full-page illustra- tions. Abstract.- Abstracts should be a summary of the basic findings, and should not exceed 2 to 3 percent of the text in length. Papers in a language other than English must be accompanied by an English abstract as well as an abstract in the language of the text. Footnotes.-Footnotes will be permitted on page 1 when it is appropriate to acknowledge grant support and to indicate change of address, etc. No footnotes will be permitted on any other page. Type footnotes on a page by themselves. Running head. -At the top of each right hand page will be printed the author(s)’ last name(s) and an abbreviated title. This running head may not exceed 55-60 characters and spaces, must be typed in capital letters, and a hyphen must separate the name(s) from the title. Tables with legends.- Prepare all tables precisely as they are to be typeset. Construct tables as simply as possible. Include legend at the top of each. Make marginal notations in the text which clearly indicate the appropriate point of insertion of each table. Note that in these instructions “graphs” are regarded as “figures,” not “tables.” The size of the printed page imposes a limit on the size of tables that can be accepted. On a normal, upright table these limits are 99 characters and spaces per line and 56 lines long, including legend. Tables printed sideways on a page can be up to 158 characters and spaces per line and 36 lines long, including legend. Sideways tables are difficult to typeset and present problems during paste-up; therefore, they will be accepted in exceptional cases only, as most tables that are “too wide” can usually be re-arranged to fit into the more desireable, upright position. Any table that exceeds the dimensions given above will not be accepted for publication. Tables must be typed double spaced throughout, and the legend must be in the following style: Table 1.- Incorporate into legend all information necessary to make the data presented clearly understandable. Do not resort to use of footnotes. Figures. -Figures are the only parts of the paper for which the author is responsible for preparing camera-ready material, and they must be done with great care. Drawings should be done with black ink on high quality, smooth, white, heavy paper. Drawings done in some manner not employing black ink (e. g., shaded pencil drawings) are difficult to print, more so than photographs. The Journal of Arachnology 9:348 When such a technique is used, the drawings should be fairly dark, have good contrast, and be well executed. The largest dimensions on the printed page to which illustrative material (with legends) will be reduced is 127 by 200 mm. Decisions as to size of originals, placement of several smaller drawings on a single page, etc., should be based upon this format. Where several smaller drawings are to appear on a single page, it is the responsibility of the author(s) to furnish an assembled “plate” precisely as it should appear in print. The “plate” must be prepared to allow for inclusion of the legend beneath the drawings. Number all drawings consecutively using Arabic numerals, do not designate illustrations with letters and never designate plates. Labelling should be done neatly, using either pressure sensitive characters or a lettering apparatus employing black ink. Photographs should have good contrast, be glossy or semi-glossy, and printed slightly larger (up to 2X) than anticipated published size. Illustrative material of any kind larger than two times the anticipated printed size will not be accepted. Larger illustrations are difficult to handle and get easily damaged or lost in the mail. Authors preparing larger illustrations should consult local photo shops and printing firms about the availability of PMT’s (PhotoMechanical Transfer) for reducing their illustrations to acceptable size. PMT’s are high contrast, direct positive-to-positive reproductions on semi-glossy paper which will also give the author(s) a good idea of how the illustrations will look when printed (The Journal of Arachnology does not provide proofs of illustrations). PMT’s can be obtained at a reasonable cost in a number of sizes, usually 8 by 10 in. or 9 by 12 in., with any amount of figure reduction desired within that space. Sending PMT’s rather than originals avoids the loss of art work in the mail, and if the PMT’s get lost or damaged they can be easily replaced. Illustrations, PMT’s, and photographs must be mounted on hard white mount- ing board (Scotch Magic Transparent Tape can be used). Illustrative material that is poorly prepared, too large, or not mounted properly will be returned to the author. The Journal of Arachnology recognizes the need and desirability to have, under special circumstances, color illustrations (including photographs). The economic resources of the Society, however, cannot provide the funds required to cover the expense of color illustrations; therefore, authors must assume full responsibility for the extra charges from the printer. Authors contemplating submission of a manuscript with color illustrations must first consult with the Editor. The number of color illustrations, and number of color separations needed must be specified in order to obtain a cost estimate from the printer. This estimate, provided free of charge, should enable the author(s) to determine whether they will be able to meet the extra expense before formally submitting the manu- script. The ability to meet the cost of color illustrations is no guarantee that they will be published. The final decision will be made by the editors based upon the recommendations of the reviewers. RESEARCH NOTES Arrange the various parts of your research notes in the following sequence: (1) mailing address, (2) title, (3) body of text, (4) by-line, (50 figure legends, tables with legends, and illustrations. Follow instructions given above for feature articles unless otherwise indicated below. Do not include foot- notes anywhere, references to grant support should be made in a statement in the body of the text. If an indication of change of address is desired, it should be included parenthetically after the credited institution as “present address.” If only a few references are cited in the text, these should be made parenthetically where they appear in the text following the style set for feature articles but omitting the title of the article cited (e. g., Jones, J. 1967. J. Arachnol. 2: 199-214). If more than a few citations are to be made, however, separate them from the body of the text and include them in their own section as in feature articles. The by-line must be typed-in paragraph form after the body of the text, or the literature cited when one is present. THE AMERICAN ARACHNOLOGICAL SOCIETY President: Jonathan Reiskind (1981-1983) Department of Zoology University of Florida Gainesville, Florida 32601 Membership Secretary: Norman I. Platnick (appointed) American Museum of Natural History Central Park West at 79th Street New York, New York 10024 Secretary: William A. Shear (1980-1982) Department of Biology Hampden-Sydney College Hampden-Sydney, Virginia 23943 President-Elect: Susan E. Riechert (1981-1983) Department of Zoology University of Tennessee Knoxville, Tennessee 37916 Treasurer: Norman V. Horner (1981-1983) Department of Biology Midwestern State University Wichita Falls, Texas 76308 Directors: Herbert W. Levi (1981-1983) Michael E. Robinson (1981-1983) Vincent D. Roth (1980-1982) The American Arachnological Society was founded in August, 1972, to promote the study of the Arachnida, to achieve closer cooperation between amateur and professional arachnologists, and to publish The Journal of Arachnology. Membership in the Society is open to all persons interested in the Arachnida. Annual dues are $20.00 for regular members, $12.00 for student members. Correspondence con- cerning membership in the Society must be addressed to the Membership Secretary. Members of the Society receive a subscription to The Journal of Arachnology. In addi- tion, members receive the bi-annual newsletter of the Society, A merican Arachnology. American Arachnology , edited by the Secretary, contains arachnological news and comments, requests for specimens and hard-to-fmd literature, information about arach- nology courses and professional meetings, abstracts of the papers presented at the Society’s meetings, address changes and new listings of subscribers, and many other items intended to keep arachnologists informed about recent events and developments in arach- nology. Contributions for American Arachnology must be sent directly to the Secretary of the Society. The Eastern and Western sections of the Society hold regional meetings annually, and every three years the sections meet jointly at an International meeting. Information about meetings is published in American Arachnology, and details on attending the meetings are mailed by the host(s) of each particular meeting upon request from interested persons. The 1982 Eastern section meeting will be hosted by William A. Shear and Hampden- Sydney College, Virginia. The 1982 Western section meeting will be hosted by William F. Rapp and the Department of Health, Lincoln, Nebraska. CONTENTS THE JOURNAL OF ARACHNOLOGY VOLUME 9 FALL 1981 NUMBER 3 Feature Articles The family Iuridae Thorell (Arachnida, Scorpiones), Oscar F. Francke and Michael E. Soleglad 233 The erigonine spiders of North America. Part 4. The genus Disembolus Chamberlin and Ivie (Araneae, Linyphiidae), A. F. Millidge 259 Effects of clearcutting on the spider community of a Southern Appalachian forest, Frederick A. Coy le 285 A comparative study of the supercontraction of major ampullate silk fibers of orb-web-building spiders (Araneae), Robert W. Work 299 The harvestman genus Liopilio Schenkel (Opiliones, Phalangiidae), James C. Cokendo Ipher 309 Observations on the natural history of Sphodros abboti and Sphodros rufipes (Araneae, Atypidae), with evidence for a contact sex pheromone. Frederick A. Coyle and William A. Shear 317 Feeding and growth of Coelotes atropos (Araneae, Agelenidae) at low temperatures, C. W. Aitchison 327 On the spider genus Odontodrassus (Araneae, Gnaphosidae), Norman I. Platnick 331 Research Notes New pseudoscorpion synonymies (Pseudoscorpionida, Chernetidae and Cheliferidae), William B. Muchmore 335 The identity of Olpium minutum Banks (Pseudoscorpionidae, Olpiidae), William B. Muchmore 337 Use of spider threads as resting places by tropical insects. Enrique J. Lahmann and Claudia Ma. Zuniga 339 Synemosyna bicolor is the female of Synemosyna americana (Araneae, Salticidae), Bruce Cutler 342 Cold survival of Argiope aurantia spiderlings (Araneae, Araneidae), Wayne A. Riddle 343 Instructions to Authors Others 346 Cover illustration, Schizomus pentapeltis (Cook), by W. David Sissom Printed by The Texas Tech University Press, Lubbock, Texas Posted on October 1981 The Journal of ARACHNOLOGY OFFICIAL ORGAN OF THE AMERICAN ARACHNOLOGICAL SOCIETY VOLUME 10 SPRING 1982 NUMBER 1 THE JOURNAL OF ARACHNOLOGY EDITOR : Oscar F. Francke, Texas Tech University. ASSOCIATE EDITOR'. B. J. Kaston, San Diego State University. EDITORIAL ASSISTANT: James C. Cokendolpher, Texas Tech University. EDITORIAL BOARD: Charles D. Dondale, Agriculture Canada. William G. Eberhard, Universidad de Costa Rica. Maria E. Galiano, Museo Argentino de Ciencias Naturales. Willis J. Gertsch, American Museum of Natural History. Neil F. Hadley, Arizona State University. B. J. Kaston, San Diego State University. Herbert W. Levi, Harvard University. Emilio A. Maury, Museo Argentino de Ciencias Naturales. William B. Muchmore, University of Rochester. Martin H. Muma, Western New Mexico University. William B. Peck, Central Missouri State University. Norman I. Platnick, American Museum of Natural History. Susan E. Riechert, University of Tennessee. Michael E. Robinson, Smithsonian Tropical Research Inst. Jerome S. Rovner, Ohio University. William A. Shear, Hampden-Sydney College. Carlos E. Valerio, Universidad de Costa Rica. Stanley C. Williams, San Francisco State University. THE JOURNAL OF ARACHNOLOGY is published in Spring, Summer, and Fall by The American Arachnological Society in cooperation with The Graduate School, Texas Tech University. Individual subscriptions, which include membership in the Society, are $20.00 for regular members, $15.00 for student members. Institutional subscriptions to The Journal are $25.00. Correspondence concerning subscription and membership should be addres- sed to the Membership Secretary (see back inside cover). Back issues of The Journal are available from Dr. William B. Peck, Department of Biology, Central Missouri State Univer- sity, Warrensburg, Missouri 64093, U.S.A., at $5.00 for each number. Remittances should be made payable to The American Arachnological Society. Change of address notices must be sent to both the Secretary and the Membership Secretary. Manuscripts for THE JOURNAL OF ARACHNOLOGY are acceptable only from cur- rent members of the Society, and there are no page charges. Manuscripts must be type- written double or triple spaced on 8.5 in. by 11 in. bond paper with ample margins, and may be written in the following languages: English, French, Portuguese, and Spanish. Contributions dealing exclusively with any of the orders of Arachnida, excluding Acari, will be considered for publication. Papers of a comparative nature dealing with chelicer- ates in general, and directly relevant to the Arachnida are also acceptable. Detailed instructions for the preparation of manuscripts appear in the Fall issue of each year, and can also be obtained from the Editor and the Associate Editor. Manuscripts that do not follow those instructions will be returned to the author(s) without the benefit of review. Manuscripts and all related correspondence must be sent to Dr. B. J. Kaston, 5484 Hewlett Drive, San Diego, California 92115, U.S.A. Work, R. W. and P. D. Emerson. 1982. An apparatus for the forcible silking of spiders. J. Arachnol., 10:1-10. AN APPARATUS AND TECHNIQUE FOR THE FORCIBLE SILKING OF SPIDERS1 Robert W. Work and Paul D. Emerson School of Textiles North Carolina State University Raleigh, North Carolina 27650 ABSTRACT Apparatus and auxiliary equipment for the forcible silking of spiders are described and illustrated. These facilitate the identification of the glandular sources of the fibers, allow for their localized isolation on a wind-up mandrel, and make possible their removal as continuous lengths or toroidal bundles for further study. Detailed descriptions are given for preferred techniques. BACKGROUND The silk fibers produced from the major ampullate gland systems of orb-web-spinning spiders have been the main subject of a continuing program of research. It is generally agreed that this pair of fibers is found in the orb web and is the essential constituent of the dragline and the trailing silk. It cannot be known who first discovered that such silk also can be forcibly drawn from immobilized spiders, but Wilder (1868) described the method. It is the normal means of securing large samples (Zemlin 1968). Work (1976) found that in such an operation minor ampullate silk fibers may also be taken inadver- tently. Subsequent papers (Work 1977a, 1977b, 1978, 1981a, and 1981b) emphasized the need for and means of differentiating between these two types of fibers, similar in some properties but quite different in others. Thus, very early in the present investigation it became imperative to develop apparatus wherewith one or the other or both fibers could be secured with certainty and do this under controlled conditions. It follows that the silk taken from spiders should be recoverable with reasonable sureness and ease. Also, it should be possible to obtain and associate the portion of a total sample with the ‘National Science Foundation Grants G.K. 33935, ENG 72-0356 and CPE 79-08905 (current) and Research Corporation Cottrell Grants of 1972 and 1979 (current) have provided funds to support the research and to develop and construct the apparatus described in this paper. 2 THE JOURNAL OF ARACHNOLOGY conditions under which it was produced, when those conditions were changed during a silking, as for example, velocity of silk withdrawal. Finally, it has been necessary to secure samples large enough for X-ray diffraction measurements and amino acid assays (which will be the subjects of a future paper). The following section (R. W. W.) will describe the method of forcible silking and will refer to the mechanical device used to the degree necessary for clarity; the next section (P. D. E.) will describe the apparatus in sufficient engineering detail so as to make possible its duplication. METHODS Although one investigator can manage the forcible silking operation, a second observer should be available if this is possible. The wind-up mandrel (in this case expandable) upon which the silk fibers are to be accumulated, and its driving mechanism are seen, A and B respectively, in Figure 1. Certain auxiliary equipment is needed, much of which is com- monly available in laboratories, and most items of which are illustrated in Figure 1 . Of these, the most essential item is a Greenough type, stereo microscope, C, equipped with a zero objective, 10X oculars and a IX to 7X zoom feature, and mounted on a cantilever arm. One, preferably two, microscope illuminators, D, are needed, as is the usual collec- tion of manipulative tools, of which a micro dissecting set (Clay-Adams), E, is preferred. A ready supply of about l”-2" lengths of narrow self-adhesive tape, F, should be at hand. While it is not imperative that carbon dioxide be available for anesthethization, (out of Figure 1 , to the left) its use reduces the hazard of injury to spiders. A bubbler, G, in the supply line provides for visual observation of flow rate, and a two way stopcock, H, on its exit side allows the gas to be directed to either of the two places where it will be needed. One of these is the “operating table”, I, being a plenum chamber with a porous plastic top surface; the other any simple glass jar, J, with a few holes punched in its metal cover. Finally, the observer will find useful a head-band mounted jeweler’s loupe, K, (3 or 4X Fig. 1. -Lay out of equipment for the forcible silking of spiders, as detailed in the text. WORK AND EMERSON-FORCIBLE SILKING 3 mag.), which may be pivoted quickly into position when needed, or raised vertically when not required, or when looking through the microscope. With items A to K made ready, the microscope is moved into position and focused, the lights brought to bear on the operating table, the forcible silk guides, L, each consisting of two staggered needles embedded in a wooden supporting rod, are positioned, as may be seen in Figure 2. The microscope body is then rotated so as to provide unimpeded access to the operating table. The spider is placed in a clean jar, J, and carbon dioxide supplied by the hose, M, is directed into the jar through one of the holes in its cover. When the spider becomes quiescent, any silk fibers entangling it and cemented to the inside surface of the container are cut free, care being taken to note the presence of trailing silk leaving its body. It is transferred, ventral side up, to the porous plate of the operating table. Any remnants of silk which will interfere with the pinioning of legs, other than the trailing silk, usually held by a fourth leg, are cut loose and removed with tweezers. Fig. 2. -“Operating table” and auxiliaries prior to pinioning of spider, as detailed in text. 4 THE JOURNAL OF ARACHNOLOGY Whether the spider is placed posterior toward or away from the observer will depend on experience with species: this investigator prefers the latter position. The same applies to the conditions for and the order of pinioning the legs with short lengths of self-adhesive tape. But, it is essential that this be done quickly, as the effect of the carbon dioxide is rapidly lost. It is here that four hands are better than two. Each leg can then be fully extended with tweezers while the cephalothorax is gently held in place, and a first strip of tape placed across the leg. Sometimes two legs can be trapped with one length of tape. But if the spider becomes suddenly and violently active during this pinioning, and appears to be in danger of damaging itself by loss of legs, it is advisable to envelop it with a loose wad of facial tissue. It can then be freed from the tape already in place, reanesthetized, and the pinioning repeated. If one of the fourth legs holds the trailing silk in its tarsal claw, it is best to pinion this one last, after cutting the silk free and taking care not to trap the end under the tape. With the spider now prevented from struggling free, it may be expedient to reposition or add additional lengths of tape to any leg that may appear to be insecure. When the effect of the carbon dioxide is dissipated, the observer will probably receive a hint as to what is to be expected. Some spiders will remain quiescent; others will struggle. It is possible that the behavior is species related \Araneus diadematus Clerck and Argiope aurantia Lucas behaved in the former manner; Neoscona hentzi (Keyserling) in the latter in the present study, for example. Each type will require the use of different methods, and the non-resisting will be considered first. If the trailing silk is obvious, it may be possible to grasp it with tweezers without swinging the microscope into position to locate it. If not, and even if under magnification no silk can be seen emerging from between the folded inward anterior and posterior spinnerets, an attempt must be made to stimulate its start. Often this can be done by gently inserting the end of a hooked micro dissection needle between the anterior spinnerets and stroking their piriform bearing surfaces. This may trigger the deposition of piriform cement which may in turn trap one or more of the major or minor ampullate fibers. Whatever is secured is drawn slowly away from the abdomen until a length is available that can be grasped with tweezers. The chances are that the observer will not know what is being carried away and brought to the mandrel. But having secured a connecting line of silk from the body of the spider to the mandrel it is now time to attempt to determine what is ready to be wound. While observing the spinneret area through the microscope, the mandrel is rotated slowly by hand to make the connecting line taut. At about 20 to 30X magnification, from one to four entities will be seen, although rarely, even a sheet of fine fibers will be found. If there are four, the two anterior will be seen to be larger that the two more posterior, indicating the presence of major and minor ampullate pairs, respectively. With three, sizes may indicate which one is missing. When there are two, the greater probability is that they are major ampullate, but this is far from being a certainty. Their presence can be confirmed by starting the motor drive and pulsing the speed of the mandrel, while watching the anterior spinnerets. If these respond by becoming more erectile and then less so, as a function of velocity of silk removal, the major ampullate spigots on these spinnerets is supplying the material. But this does not prove that the minor ampullate pair is not also being taken, since each of the pair may have become attached in line contact with its corresponding major. With some species the mandrel speed may be increased to a point where the sources of the fibers can be seen with very strong vertical illumination and 50 to 70X magnification. Whatever the situation, when in doubt it is best to make use of the flexibility of the silking equipment to isolate the fibers being secured. But before this step is described it is necessary to return to the problems posed by the spiders which react negatively to pinioning and forcible silking. WORK AND EMERSON-FORCIBLE SILKING 5 No facile descriptions can be given as to what an investigator should best do to secure the desired sample in such a situation. Sometimes the difficulties can be solved by sheer presistence. If not, the spider may require complete anesthetization. But this is not an easy way out. In order to start the forcible silking of such a spider, the silk, as yet of undefined origin, must be where it can be grasped with tweezers. The inert animal does not respond to the stroking of the piriform spools. Furthermore, as has been already reported (Work 1976), the physical properties of fibers taken from a fully anethetized spider may differ from those otherwise secured. Sometimes a compromise condition may be achieved by the use of carbon dioxide on a periodic or pulsing basis. This is aided by the use of a specially constructed cell. Carbon dioxide may be administered by flooding the plenum chamber with it and allowing it to rise through the porous plate to which the spider is pinioned. Being heavier than air, this gas then tends to surround the spider. A glass cell inverted over it helps to retain the gas and gives the observer additional control over conditions. But the construc- tion of such a device, although simple, necessitates the skill of a glassblower. It consists of a piece of glass tubing of diameter somewhat larger than the leg span of the pinioned spider. This is cut to a length greater than the distance between the porous plate and the top most point of the spider’s body. At one place this cylinder is cut from one end almost to the other, making a slot about 1 mm wide, which is then fire polished. A piece of optical glass, as for example a double size microscope slide is then cemented to the upper end of the cell and the excess trimmed away with a diamond saw. This cell can be placed over the spider after the silk is fastened to the mandrel, and then observations can be made by means of the microscope through the optical glass, with the silk being led to the mandrel via the slot. Such a cell, N, is partially hidden, but is in Figure 1 . The silk having been started from the spider by one means or another, secured to the mandrel and then tentatively identifed as to source, can not be collected. Guide pairs of needles may be spaced as desired in a small wooden dowel rod, placed parallel and adjacent to the mandel. Each fiber entity desired as a separate sample is then led between the pair of guide needles which will place it on a preselected position on the mandrel. The capacity of this last to be moved horizontally at a uniform rate as it rotates or remain in a fixed position, allows for the primary samples to be placed as a helix or in a piled-up bundle, or the former may preceed or follow the latter, as desired. In Figure 3 a pinioned Argiope argentata (Fabricius) is seen from which major and minor pairs of ampullate silk have been started to be wound on an expandable mandrel. Arrow, P, indicates the band of both fibers (to be discarded later) placed at the beginning of the operation, during which identification had been made. The silking was then stopped, the minor pair transferred between the left guide needles and the major pair allowed to remain between the right guide needles. The silking was then restarted as a very slow (1 cm/sec) rate, and at the same time the dowel was moved slowly from left to right. This placed a helix of each type of silk on the mandrel, from which secondary samples later could be rewound for positive identification or any other study that might be desired. The helices are identified by brackets Q and R, respectively. At the right side of each helix the large samples have been started. Figure 3 also illustrates the use of a piece of V* dowel rod, S, to raise and immobilize the spider’s abdomen. The 2-way stopcock of the carbon dioxide system is seen in the background. It is essential that the primary sample be a helix if the investigator wishes to make a positive identification of a bundle to follow, or if a number of primary samples are needed for individual study. The former, of course, is essential if an X-ray diffraction or 6 THE JOURNAL OF ARACHNOLOGY amino acid investigation is involved. But if only a smaller amount is required, then the expandable type of mandrel is best replaced with a simple cylindrical one (painted black for contrast). This allows for the advancing mechanism of the machine drive to produce a uniformly spaced helix without requiring the investigator to make it in a less satisfactorily controlled manner. The transfer of samples from the helix can be started at any selected point by first placing tiny tabs of self-adhesive tape on the primary sample at each side of that place. It is then cut with a micro scalpel between the tabs, one of which will remain in place to prevent a loose end from interfering with the backwinding. The other tab, with the end of the silk sample adhering to it, is grasped with tweezers and as the mandrel is back rotated by hand, the transfer can be made. In the event that the sample is broken or the end being manipulated pulls free of the tab, these being not unusual happenings, the lost end can be found or a new one started from the helix, which is a virtual impossibility from a bundle. It has been found that with those spiders which do not resist forcible silking, winding can be done at about 3 m/min. for periods of ten to twenty minutes. During this time the spider appears to be capable of supplying a continuous flow of progenitive polypeptide and in turn allow it to be converted to silk fibers by the drawing action of the forcible silking operation. (As a first approximation, under these conditions a mature female Araneus diadematus Clerck, furnishing a pair of major ampullate fibers, each of 3/im diameter, will supply slightly less than 0.06 mg/min. of primary sample). Throughout this entire silking the process must be observed by means of the microscope. If and when there is any disturbance in the spinneret area, chiefly the back and forth rubbing together of the piriform bearing surfaces of the anterior spinnerets, it is necessary to stop the mandrel immediately. If there is any question in the observer’s mind that a change could Fig. 3.- Spider, A. argentata pinioned ventral side up, with major and minor ampullate pairs of silk fibers being wound on the expandable mandrel. WORK AND EMERSON-FORCIBLE SILKING 7 have taken place, it is advisable to move a continuing sample to an unused section of the mandrel. Since an observer must remain watching the operation, with one hand on the control switch of the motor driven mandrel for as long as twenty or more minutes, it is highly desirable that here, as in the pinioning step, a second person should be available to alternate between note taking and observing. A bundle of fibers wound onto an untapered mandrel cannot be removed, except by cutting it free. This may be a satisfactory solution in some cases. But a toroidal bundle of a continuous fiber or pair of fibers is essential for certain operations, and in any case, may be manipulated with ease, as compared with the same bundle that has been cut at one point. An expandable mandrel makes the former possible. In operation the slotted section of the mandrel is expanded before the bundle sample is to be wound on it. After it has been accumulated and its lead-in backwound, the expanding plug is removed (a wrench on it and a second on a flattened section of the mandrel will be required). To facilitate removal of the sample and aid in its subsequent manipulation, it has been found to be useful to provide it with “handles.” These are conveniently made from continuous fila- ment nylon sewing thread. Cotton or any other non-continuous filament thread should not be used, since these may provide fiber fragments as contaminants. Colored nylon thread provides contrast and, if desired, species can be identified by using a different color for each. To make the handle, the end of the nylon thread is passed into the aperture normally occupied by the expanding plug (now removed), up through the slot required to provide for expansion, over the bundle and knotted in a convenient loop. A double knot should be used since nylon knots are apt to slip. Two of these loops, are ample to allow for handling without the need for the investigator to touch the sample. One of the loops may be allowed to have a long end, to which an identifying label may be attached. With the loops in place, the bundle is urged toward the outer end of the mandrel by means of a hooked micro dissecting needle, again making use of the slots as openings. It is necessary to do this by very small individual movements of the bundle, going around it from slot to slot. Any attempt to force it will run the hazard of fiber breakage and subsequent tangling and snarling. At the very end of the removal the nylon thread loops are grasped and separated, at 180° to each other, by one operator. The other operator completes the shift of the fiber bundle off the mandrel, while the one holding the loops keeps them far enough apart to prevent the bundle from snapping into a convoluted “muff’ at the instant of relaxed strain. A common laboratory glass desiccator, without desiccant, provides a useful means of storing the samples, to prevent contamination by particulate matter in the air. APPARATUS An apparatus for forcible silking consists of three essential elements: 1) a variable speed drive, 2) a rotating take-up mandrel, and 3) means for traversing the filament as it is wound. Desirable features include compactness, easy accessibility and versatility. As illustrated in Figure 4 the apparatus consists of a 10” square base which supports front and side panels 10” high. Although a totally enclosed cubical box could be employ- ed, the open structure facilitates access to the drive mechanism as necessary. Mounted on the front panel is a Minarik model SL32 speed control designed specifically for operation with a Bodine-34 115 volt DC motor. The control provides two speed ranges which are infinitely variable by means of the centrally mounted control knob. Although normal 8 THE JOURNAL OF ARACHNOLOGY operation is forward, with the spindle moving clockwise, the control provides reverse rotation capability. A run-stop switch located at the lower right corner of the control allows the operator to start and stop the spindle by touch while closely observing the spider through the microscope. Protruding through a 1 Vi” diameter opening in the side panel is the spindle-mandrel assembly onto which the silk filaments are wound. Two types of mandrels are used, one having an expandable portion at the end, and the other being a plain cylinder with a black anodized surface. Both mandrels are l” diameter and 4 " long. As illustrated in Figure 5 the motor shaft is connected to a countershaft by means of a %” pitch chain. Since fractional horsepower dc motors usually do not rotate smoothly at very low speeds a two-stage speed reduction of 4 to 1 from motor to spindle shaft is provided by suitable sprockets. A 20 tooth sprocket on the motor shaft drives a 40 tooth sprocket on the countershaft. The 10 tooth sprockets on the end of the countershaft drive two sprockets mounted on the spindle shaft. One of these is a 21 tooth sprocket which is free running on the spindle shaft. The other sprocket has 20 teeth and is locked on the spindle shaft by means of a set screw. A flat, milled into the spindle shaft, permits the set screw to be tightened without deforming the cylindrical surface of the shaft. Inserted in the hub of the 21 tooth sprocket are two 1/8” diameter steel rods which extend into coresponding holes in the mandrel. The end of the spindle shaft is threaded, (3/8-24-NF), and the mandrel screws onto the shaft as shown. In operation, the spindle shaft is direct driven by the motor drive when the 20 tooth sprocket is locked to the spindle shaft. The free-running 21 tooth sprocket rotates at 95% of the speed of the spindle shaft. Since the free-running sprocket drives the mandrel 5% faster than the speed of the spindle shaft, the mandrel is thereby gradually advanced outward on the threaded portion of the spindle shaft. For each turn of the spindle shaft the mandrel advances 5% of the lead of the screw thread on the spindle shaft. As a result, the lead of the spider silk helix is 5% of the 3/8-24 NF thread, or 0.05 x 0.0417”, which is about 0.002”. This makes it possible to wind an evenly spaced helix approximately 40 PLAIN MANDREL Fig. 4. -Apparatus for the forcible silking of spiders. Fig. 5.— Drive mechanism of the apparatus. WORK AND EMERSON-FORCIBLE SILKING 9 m long onto each axial inch of mandrel, which at about 2 m/min, will require somewhat more than 13 minutes. If conditions are changed during silking, it is only necessary to move the silk guide needles (Fig. 2, L) to the left so as to provide an obvious gap to denote the new situation. To set the mandrel in starting position it is merely necessary to loosen the set screw in the hub of the 20 tooth sprocket so that the spindle shaft may be screwed back into the mandrel by means of the hand crank, while restraining the mandrel from rotation. The set screw should then be re-tightened. In the event that it is desired to wind a silk bundle, rather than a helix, the set screw may be loosened and there will then be no mandrel advance. In this condition, drive to the mandrel will be through the 21 tooth sprocket and the 20 tooth sprocket will be free-running. To provide the capability for diameter reduction of the mandrel, thus facilitating removal of tightly wound bundles, an expandable mandrel is desirable. This device is fitted with a Vi" NPT pipe plug threaded into the outer end of the mandrel, which is slotted to provide for the expansion caused by the tapered pipe plug. Prior to winding, the plug is screwed all the way into the mandrel. When winding is complete the mandrel can be prevented from rotating by being held in the jaws of a wrench, using the flats, while another wrench is employed to remove the pipe plug thus allowing the mandrel to contract. DISCUSSION Although the apparatus described in this paper was developed and has been used for the forcible silking of spiders, its versatility may well be adapted to the silking of other silk-producing animals. In this connection, an axiom from the field of macromolecular chemistry of linear, fiber-forming molecules must be kept in mind. It is known that strong fibers cannot be made by simple extrusion. Molecular segments must be oriented by stretching (technically called drawing) the macro structure during some phase of the production of the fibers. An example would be the drawing out of the progenitive polypeptide by the side to side wagging motion of the larva of Bombyx mori, as it forms its cocoon. It has been brought to the attention of one of the authors (RWW by Ms. Lottie Spainhour) that when B. mori is in the “cocoon-ready” stage, silk may be forcibly drawn from it. The same may be possible with other members of the same order. If such would be the case, using the described apparatus, it might be possible to secure samples of silk under controlled conditions in the laboratory, rather than from unravelled cocoons or in the field. LITERATURE CITED Wilder, B. G. 1868. On the Nephila plumipes, or Silk Spider. Proc. Amer. Acad. Arts Sci., 7:52-57 . Work, R. W. 1976. The force-elongation behavior of web fibers and silks forcibly obtained from orb-web-spinning spiders. Text. Res. J., 46:485-492. Work, R. W. 1977a. Mechanisms of major ampullate silk fiber formation by orb-web-spinning spiders. Trans. Amer. Micros. Soc., 96(2): 170-189. Work, R. W. 1977b. Dimensions, birefringences, and force-elongation behavior of major and minor ampullate silk fibers from orb-web-spinning spiders- The effects of wetting on these properties. Text. Res. J. 47:650-662. Work, R. W. 1978. Mechanism for the deceleration and support of spiders on draglines. Trans. Amer. Micros. Soc., 97(2): 180-191. 10 THE JOURNAL OF ARACHNOLOGY Work, R. W. 1981a. Web components associated with the major ampullate silk fibers of orb-web- building spiders. Trans. Amer. Micros. Soc., 100: Work, R. W. 1981b. A comparative study of the supercontraction of major ampullate silk fibers of orb-web-building spiders J. Arachnol., 9:299-208. Zemlin, J. C. 1968. A study of the mechanical behavior of spiders’ silks. Tech. Rep. 69-29-CM, AD 684333, U.S. Army Natick Laboratories, Natick, Massachusetts. Manuscript received July 1980, revised September 1980. Mahnert, V. 1982. The pseudoscorpion genus Corosoma Karsch, 1879, with remarks on Dasychernes Chamberlin, 1929, (Pseudoscorpiones, Chernetidae). J. Arachnol., 10:11-14. THE PSEUDOSCORPION GENUS COROSOMA KARSCH, 1879, WITH REMARKS ON DASYCHERNES CHAMBERLIN, 1929 (PSEUDOSCORPIONES, CHERNETIDAE) Volker Mahnert Museum d’Histoire naturelle Case postale 284, CH-121 1 Geneve 6, Suisse ABSTRACT After re-examination of the type specimen of Corosoma sellowi Karsch, the genus is transferred from the chernetid subfamily Lamprochernetinae to the Chernetinae (no tactile seta on tibia of leg IV, flagellum of three blades). Probably it is closely related to Dasychernes Chamberlin. Some external adaptive characters suggest a way of life similar to that of Dasychernes, which is known to inhabit nests of meliponine bees. INTRODUCTION In 1879 Karsch described the new genus and species Corosoma sellowi, based on a single dried specimen collected at St. Paul (Sao Paulo), Brazil, and mentioned the follow- ing characters for the genus: carapace very broad at the posterior margin and narrowed anteriorly, with two distinct transverse furrows; abdomen broader than long, the tergites divided, the margins of the scuta forming nearly a right angle at the median line; no eyes; sclerotic parts shining (smooth), setae numerous (“Behaarung. . .ziemlich dicht, gelb- grau”). To Karsch the genus seemed to be nearest related to the genus Garypus L. Koch, 1873. In 1930 Beier gave a complementary description, illustrating the chelicera, the left pedipalp, and legs I and IV, and considered Corosoma Karsch a junior subjective synonym of Lamprochernes Tomosvary. He revised this point of view in his world monograph (1932) and placed Corosoma as an uncertain genus in the Lamprochernetinae. Roewer (1936) illustrated the chelal finger of C. sellowi showing a number of “Chitinkegel” on the inner and outer sides of the fixed finger. He followed Beier (1932) in accepting the uncertain position of the genus. Dasychernes was proposed by Chamberlin in 1929 for the new species inquilinus from nests of meliponine bees in Colombia, the most important characters of the new genus being: carapace almost smooth, setae non-denticulate; garypoid in form; two strongly developed transverse furrows; leg IV without tactile seta, claws and subterminal seta simple; tergites distinctly hairy especially laterally and posteriorly, divided; setae numerous and evenly distributed over the scuta; flagellum of three blades. This somewhat fragmentary description was supplemented subsequently by the author in 1931 with figures of cephalothorax, flagellum, galea, pedipalp, trichobothrial pattern, male genitalia, and some other morphological details. 12 THE JOURNAL OF ARACHNOLOGY The descriptions of the two species suggest a degree of similarity if not a synonymy of Coro soma and Dasy cherries. To establish the identity and taxonomic position of Corosoma sellowi Karsch, the type specimen is redescribed below. Corosoma sellowi Karsch Corosoma sellowi Karsch 1879:95. Lamprochernes sellowi: Beier 1930:298-300, figs. 9-12. Corosoma sellowi:: Beier 1932:105 (“unsicheres Genus”); Roewer 1936:56, fig. 31; 1937:291 (“unsicheres Genus”). Type specimen.— Adult, dried and pinned; pedipalp, chelicera and legs I and IV mounted on four microscope slides (Zoological Museum Berlin, ZMB no. 880); St. Paul, Brasilien, Sellow lg. Description.— Sex indeterminable, genital area destroyed by the pin; carapace smooth, reddish brown, apparently with numerous setae, strongly trapezoid in form; two indis- tinct eyespots; two distinct transverse furrows, the subbasal clearly nearer to the posterior margin than to the submedian furrow; tergites smooth, I to X divided, with numerous evenly distributed long and finely dentate setae (clustered at the lateral margin); chelicera with 7 setae on palm, movable finger with 4 galeal setae (two of them broken), and a small tooth-like subapical lobe; galea broken (probably multi-branched); serrula exterior of approximately 50 lamellae, flagellum of three long blades (the distal one dentate); fixed finger with 6 pointed teeth and two distal granular ones; coxae smooth, with numerous simple and long setae. Setae of pedipalps and legs long and simple; pedipalp smooth; trochanter without protuberance; femur 3.2x as long as broad; tibia 2.7x;hand cylindrical, elongate, with pedicel 2.3x as long as broad and 1.53x longer than finger; chela with pedicel 3.7x, without pedicel 3.3x as long as broad; each finger with approxi- mately 46 small teeth, fixed finger with 16 external and 1 internal (distal) accessory teeth, movable finger with 9 external accessory teeth, none medially; nodus ramosus situated at the level of trichobothrium st\ sense spot area on fixed finger well developed, fingers not gaping. Trichobothria: est and ist in basal third of fixed finger, only et and it (situated at the same level) in the distal half of the finger. Leg I: basifemur 1.2x as long as broad, telofemur 2. Ox as long as broad and 1 .43x longer than basifemur; tibia 2.8x; tarsus 5. Ox; claws simple. Leg IV: all joints with numerous setae, femur 2.9x, tibia 4.6 x, tarsus 5.5x as long as broad; no tactile seta on tibia or on tarsus; subterminal setae slightly curved, simple. Legs as figured by Beier (1930). Measurements (in mm).— Pedipalp: femur 1.88/0.59; tibia 1.71/0.62; hand with pedi- cel 1.65/0.70, length of pedicel 0.27, length of movable finger 1.08; chela length with pedicel 2.61, without pedicel 2.35; leg I: basifemur 0.76/0.63; telofemur 1 .09/0.54; tibia 1.06/0.38; tarsus 1.04/0.21; leg IV: femur (total) 1.90/0.65; tibia 1.55/0.34; tarsus 1.21/0.22. The differences in measurements given by Beier (1930) and here may be explained by the fact that Beier himself did not check the type specimen but published measurements taken by Roewer; there are some obvious discrepancies between the mea- surements given and the drawings (e.g. length of tibia). The re-examination of the type specimen of Corosoma sellowi Karsch revealed two facts: Corosoma has to be considered as a good genus, and it should be placed not in the Lamprochernetinae but in the Chernetinae (absence of tactile setae from tibia and tarsus MAHNERT-PSEUDOSCORPION GENUS COROSOMA 13 of leg IV). Within this sub-family it shares quite a number of characters with the genus Dasychernes Chamberlin: garypoid form of cephalothorax, which is nearly smooth; numerous setae evenly distributed on tergites; no tactile setae on leg IV. The two genera differ in the number of galeal setae on the movable finger of the chelicera (4 in Coro soma, 1 in Dasychernes ), in the trichobothrial pattern on the chelal fingers ( est nearer to esb than to et in Corosom, but nearer to et than to esb in Dasychernes ; sb a little nearer to b than to st in Corosoma, much nearer to b than to st in Dasychernes), and in the general shape of the pedipalp (cf. Chamberlin 1931, fig. 27C). Unfortunately, I was not able to check the type specimens of Dasychernes inquilinus Chamberlin, but my colleague, Dr. W. B. Muchmore, University of Rochester, had examined the specimens and kindly sent me his worksheets. Since no measurements are mentioned in the original description, we give here the principal ones: d-holotype and 9-allotype (JC-439. 01002, JC-439. 01001) (in mm). -Length of carapace of 6 1.98 (9 1.78). Pedipalps: femur 1.61/0.58 (1.55/0.49), tibia 1.48/0.60 (1.36/0.49), chela (with- out pedicel) 2.67/0.78 (2.44/0.66), hand (without pedicel) 1.11/0.80 (1.04/0.65), length of pedicel 0.31 (0.23), length of movable finger 1.70 (1.52). Leg IV: femur (total length) 1.77/0.46 (1.74/0.46), tibia 1.34/0.28 (1.27/0.24), tarsus 1.17/0.21 (1.12/0.20). Female spermatheca consists of a pair of short, apically slightly enlarged tubes. Figs. 1-5 .—Corosoma sellowi Karsch; 1, habitus; 2, left chelicera; 3, 4, pedipalp; 5, trichobothrial pattern on chelal fingers. Scales in mm. 14 THE JOURNAL OF ARACHNOLOGY Conclusions.— The pseudoscorpion genera Corosoma Karsch and Dasychernes Chamberlin both belong to the Chernetinae and are differentiated from one another mainly by trichobothrial pattern and number of galeal setae. The presence of numerous setae on cephalothorax and tergites in both genera may by explained by convergence and suggests that Corosoma, like Dasychernes , inhabits nests of (meliponine?) bees. The affinities of the two genera can be ascertained only by study of more material in future. ACKNOWLEDGEMENTS I am greatly indebted to Dr. M. Moritz, Zoological Museum of Berlin, for giving me the opportunity to check the type specimen of Corosoma sellowi Karsch, and to Dr. W. B. Muchmore, University of Rochester, for sending me his notes on Dasychernes inquilinus Chamberlin and for his linguistic help. LITERATURE CITED Beier, M. 1930. Die Pseudoskorpione der Sammlung Roewer. Zool. Anz., 91:284-300. Beier, M. 1932. Pseudoscorpionidae II. Subord. C. Cheliferinea. Tierreich, 58:xxi + 294 pp. Chamberlin, J. C. 1929. Dasychernes inquilinus from the nest of meliponine bees in Colombia (Arachnida: Chelonethida). Ent. News, 40:49-51. Chamberlin, J. C. 1931. The arachnid order Chelonethida. Stanford Univ. Publ., Biol. Sic., 7(1): 1-284. Karsch, F. 1879. Zwei neue Arachniden des Berliner Museums. Mitt. Miinchener ent. Ver., 3:95-96. Roewer, C. Fr. 1936. Chelonethi oder Pseudoskorpione. Bronn’s Klass. Ordn. Tierreichs, 5, IV, 6(1): 1-160. Roewer, C. Fr. 1937. Chelonethi oder Pseudoskorpione. Bronn’s Klass. Ordn. Tierreichs, 5, IV, 6(2): 161-320. Manuscript received November 1980, accepted December 1980. Jass, J. P. 1982. Regression estimate of population size for the crab spider Philodromus cespitum (Araneae, Philodromidae). J. Arachnol., 10:15-18. REGRESSION ESTIMATE OF POPULATION SIZE FOR THE CRAB SPIDER PHILODROMUS CESPITUM (ARANEAE, PHILODROMIDAE) Joan P. Jass Invertebrate Zoology Milwaukee Public Museum Milwaukee, Wisconsin 53233 y' ABSTRACT A field population of Philodromus cespitum (Walckenaer) was used to test Soms’ (1978) regression method of analyzing data obtained by removal sampling to obtain a population size estimate. The population density at the study site, as determined by sweep netting, was 1.25/m2 on June 18, 1978. INTRODUCTION Philodromus cespitum (Walckenaer) is a grass dwelling spider in the family Philo- dromidae. P. cespitum is a holarctic species with a broad American range (Dondale and Redner 1976). This study involves estimation of the population size of P. cespitum in a grassy parkland in Milwaukee County, Wisconsin. It represents an attempt to determine whether the sweep netting technique combined with Soms’ (1978) regression method of analyzing data obtained by removal sampling as described here could be of value in estimating the size of spider populations. MATERIALS AND METHODS The locality where the study was conducted was an unmown area on the Milwaukee River Parkway in Milwaukee County. The vegetation consisted largely of grasses with some forbs intermixed. Scattered sparsely at the site were large sugar maple trees ( Acer saccharum Marsh). That this is typical Philodromus cespitum habitat is supported by reference to Putman (1967), who describes the habitat of this species as “grassy areas, especially near trees” (p. 624) in peach orchards in the Canadian Niagara Peninsula. Also, in contrast to the other Philodromus species studied by Putman, P. cespitum was not frequently found on trunks and branches of the nearby trees. A series of sweep-netted samples with collection numbers high enough for application of statistical analysis to obtain a population estimate (at least 20 philodromids in the first sample) was acquired on June 18, 1978. The methodology followed to collect a series of 16 THE JOURNAL OF ARACHNOLOGY samples for use in estimating population size was similar to that of Menhinick (1963). A 15.24 X 15.24 meter square was roped off at the study site, an area equaling 232.5 m2 . Within this square, I used a 30 cm diameter sweep net with a 1.4 m handle, making 150 sweeps per sampling. The pattern of movement within the square was such that five 3.048-meter wide lanes were swept in 30 sweeps each. After the 150 sweeps had been made, the contents of the net were emptied into a plastic bag and poisoned with ethyl acetate. Spiders were removed and placed in 70% isopropanol. For the purpose of apply- ing the regression method of Soms (1978) to obtain a population estimate, a series of six successive samplings were made. The series of numbers representing the individuals caught in the successive samplings was then subjected to the statistical analysis. The series of numbers representing the individuals caught in the successive samples for each date were analyzed using the regression method of Soms (1978) where: p, probability of capture = 1 - q, q = nnj , n. = the number of individuals caught in sampling i, k - 1 i = (i - — )/(k(k2 - 1)/12), k equaling the number of sampling periods, s~2, the variance of the probability of capture - (q2/fn) 2c-2 /p^ m, the population estimate = 2n./(l - q ), 1 s^2 , the variance of the population estimate - - Ak mq l-qk and the 100 (1 - a)% confidence intervals for p and m are p ± z n sA and r a/ 2. p m ± z s~ a/ 1 m where z ^ is the 100 a/2 upper percentile of the standard normal. RESULTS 1 + k2 qk k Cj2 2 — l-qk 1 Pj The June 18 data set has k = 6 and (ni , n2, n4 , n5 , n6) - (42, 28, 35, 21, 24, 22) k with 2 n. = 172. The regression method of Soms (1978) gave an estimated population 1 1 size of 335 ± 89.8. This is a density of 1.44/m2 for that date. The 95% confidence interval for the June 18 estimate is 155.4 to 514.6. The 95% confidence interval for the June 18 density is 0.668 to 2.21/m2. This method of analysis gives a probability of J ASS- REGRESSION ESTIMATE OF SPIDER POPULATION SIZE 17 Table 1. -Estimators resulting from the regression technique applied to a population of Philo- dromus cespitum (Walckenaer) in Milwaukee County, Wisconsin June 18, 1978. Estimators for k=6 18 June 1978 p (probability of capture) 0.113 s.d. (standard deviation) 0.041 m (population estimate) 335 s.d. (standard deviation) Test of Fit 89.8 Z 3.48 d.f. (degrees of freedom) 0.253.48)<0.50, for d.f. =4 4.03 Individuals per Sq. Meter 1.44 95% Confidence Interval 0.668-2.21 Estimators for pooled k=2 m (population estimate) 290 s.d. (standard deviation) 63.9 Individuals per Sq. Meter 1.25 95% Confidence Interval 0.698-1.80 capture estimate of 0.113 ± 0.041. The 95% confidence interval for the probability of capture is 0.031 to 0.195. The test of fit, Z, for June 18 is 3.48 with 4.03 degrees of freedom. On the X2 tables, this test of fit figure gives the following: for June 18, 0.25 < P(X2 > 3.48)< 0.50. Thus the observed values correspond to those expected. The low value for the probability of capture estimate for the June 18 data set indicates some degree of bias. It is therefore desirable to pool the n .., using k = 2. After this was done, the estimate of m was 290, with s.d. 63.9. This new estimate gave a less biased density of 1.25/m2. The calculation of 95% confidence intervals for this pooled June 18 data set gives an interval of 162.2 to 417.8 for the population estimate and an interval of 0.698 to 1.80/m2 for the density. Zippin (1956) has pointed out that the removal method, to be reasonably precise, requires a coefficient of variation (c.v. = standard error/estimate x 100) of 30% or less. For the June 18 data, c.v. - 63.9/290 x 100 = 22%, giving acceptable precision. On June 18, 59% (172/290) of the spiders were removed from the population. Results are also shown in Table 1 . DISCUSSION No fully satisfactory method exists for estimating spider population density. Casual observations in the field may show a certain species to be relatively abundant, but obtaining a quantitative statement of abundance requires techniques that often were developed for larger and/or more easily enumerated organisms. A review of various methods and their limitations is given by Southwood (1968). Mark-recapture is one of the popular methods of making animal population size esti- mates. The use of mark-recapture on spider populations, however, has more often been to investigate features such as range size (Hallander 1967) and niche relationships (Kuenzler 1958) rather than to estimate population size. The difficulty of handling smaller spiders without injuring them also tends to confine this sort of method to larger species like the wolf spiders (Lycosidae) that were the subjects of the Hallander and Kuenzler studies. 18 THE JOURNAL OF ARACHNOLOGY Also, since this method requires the release of the captured individuals, other potential studies requiring the measurement of spiders in the sample would not be as feasible. Another technique that has been used by animal ecologists is that of removal trapping. Removal trapping entails a series of sampling periods. At each sampling period animals are removed so that there is a successive reduction in population. A regression analysis of the sequence of numbers of animals obtained during this series of sampling periods enables one to obtain an estimate of the size of the population in the study area. Papers dealing with the statistical analysis of data collected by the removal method are: Carle and Strub (1978), Moran (1951), Soms (1978) and Zippin (1956, 1958) and pages 44-50 of Otis et al (1978). There can be several different approaches to the analysis of the data gained by removal sampling. The technique used by Menhinick (1963) to arrive at a population estimate is a graphical one. A. P. Soms has subsequently devised a computationally simpler regression technique, one that is based on the limiting distribution of the multinomial, and a short computer program for arriving at the appropriate regression estimators. Soms’ (1978) alternative method also avoids statistical deficiencies inherent in other approaches such as that based on the maximum likelihood and allows for pooling when sample sizes are small. Besides a population size estimate and its standard error, Soms’ method gives the estimated probability of capture and its standard error, allowing for calculation of 95% confidence intervals, and includes a test of fit. ACKNOWLEDGMENTS I especially acknowledge the assistance of A. P. Soms in statistical analysis, C. D. Dondale in the identification of specimens and also critically reviewing the manuscript, A. M. Young in initiating the 1977 Milwaukee County spider survey, C. M. Weise in critically reviewing the manuscript, and reviewers for The Journal of Arachnology for their com- ments on earlier versions of this manuscript. LITERATURE CITED Carle, F. L. and M. R. Strub. 1978. A new method for estimating population size from removal data. Biometrics, 34:621-630. Dondale, C. D. and J. H. Redner. 1976. A review of the spider genus Philodromus in the Americas (Araneida: Philodromidae). Canadian Entomol., 108:127-157. Hallander, H. 1967. Range and movements of the wolf spiders Pardosa chelata (O. F. Muller) andP. pullata (Clerck). Oikos, 18:360-364. Kuenzler, E. J. 1958. Niche relations of three species of lycosid spiders. Ecology, 39:494-500. Menhinick, E. F. 1963. Estimation of insect population density in herbaceous vegetation with empha- sis on removal sweeping. Ecology, 44:617-621. Moran, P. A. 1951. A mathematical theory of animal trapping. Biometrika, 38:307-311. Otis, D. L., K. P. Burnham, G. C. White, and D. R. Anderson. 1978. Statistical inference from capture data on closed animal populations. Wildlife Monographs, 62:6-135. Putman, W. L. 1967. Life histories and habits of two species of Philodromus (Araneida: Thomisidae) in Ontario. Canadian Entomol., 99:622-631. Soms, A. P. 1978. Simplified point and interval estimation for removal trapping. Proc. 24th Conf. Design Exp. Army Res. Dev. Testing: 143-159. Southwood, T. R. 1968. Ecological methods with particular reference to the study of insect popula- tions. 1st ed. revised. Chapman and Hall, London. Zippin, C. 1956. An evaluation of the removal method of estimating animal populations. Biometrics, 12:163-189. Zippin, C. 1958. The removal method of population estimation. J. Wildlife Mgmt. 22:82-90. Manuscript received October 1980, revised December 1980. Randall, J. B. 1982. Prey records of the Green Lynx spider, Peucetia viridans (Hentz) (Araneae, Oxyopidae). J. Arachnol., 10:19-22. PREY RECORDS OF THE GREEN LYNX SPIDER, PEUCETIA VIRIDANS (HENTZ) (ARANEAE, OXYOPIDAE) John B. Randall1 Entomology and Nematology Department University of Florida Gainsville, Florida 32611 ABSTRACT Sixty-six prey items representing six orders, 24 families and 30+ species were collected directly from feeding green lynx spiders, Peucetia viridans (Hentz). Prey items were identified and subjectively evaluated as to their harmful to beneficial effects in order to gauge the impact of P. viridans as a predator in the agroecosystem. INTRODUCTION The role spiders play in the natural biological control of agricultural pests has received limited investigation (Huffaker and Messenger 1976). The four important roles of spiders in the agroecosystem were outlined by Whitcomb (1973) and include: a) spiders prey on destructive insects; b) spiders serve as food for other predators; c) since spiders tend to be general feeders, they are enemies of beneficial insects; and d) spiders compete with insect predators for prey. Dondale (1958) and Putman (1967) reported, that in orchards, mem- bers of the Salticidae, Thomisidae and Theridiidae were the most numerous spider species. Yeargan and Cothran (1984) indicated that in alfalfa the Lycosidae, Erigonidae, and Tetragnathidae were the dominant spider families. Prey records are necessary in analyzing the importance of spiders as predators in an agroecosystem. Previous prey investigations have indicated only that spiders are poly- phagous and that many different insects make up the spider’s diet. Prey investigations under natural conditions (Robinson and Robinson 1970) and laboratory conditions (Eason and Whitcomb 1965, Peck and Whitcomb 1970, Turnbull 1965, Whitcomb and Eason 1967) have not been able to prove or disprove that spiders regulate prey popula- tions, or that they would be effective biological control agents. Turner (1979) collected 189 prey items of Peucetia viridans in a dry coastal sagescrub area of California and reported species of Hymenoptera, particularly Apis mellifera Latreille, represented the greatest number of prey items, 41% of the prey collected. The second most numerous prey species belonged to the Diptera (15%) followed by the Present address: #7 TwaddelMill Road, Wilmington, Delaware 19807 20 THE JOURNAL OF ARACHNOLOGY Lepidoptera (15%), Hemiptera (9%), Orthoptera (8%), Araneae (7%) and Coleoptera (4%). Whitcomb (1966) also reported that in Arkansas P. viridans seized large numbers of A. mellifera as well as other hymenopterous species. Unlike web-weavers, hunting spiders like the green lynx are not restricted to prey that become tangled in a snare. P. viridans builds no snare and can be found moving about on vegetation ready to pounce on its prey. This investigation was conducted not only to add to the list of known prey of the green lynx spider, but to evaluate how harmful or beneficial those prey were thereby documenting the effect this spider may have as an economically important predator. METHODS Green lynx spiders were observed in the field from March 1974 to September 1977 at incidental locations throughout Florida. Sixty-six specimens with prey in their grasp were collected. Whether prey was actually consumed by the spiders was not documented since that information was irrelevant to this investigation. Prey specimens were preserved in alcohol prior to being sent to staff entomologists at the Florida Division of Plant Industry (D.P.I.), Gainesville, Florida for identification. The D.P.I. entomologists later subjectively evaluated the prey specimens they had identified by qualifying them as harmful or bene- ficial on a scale of -3 (most harmful) to +3 (most beneficial). RESULTS AND DISCUSSION Results of the identification and evaluation of the 66 prey items collected from P. viridans (representing six orders, 24 families and 30+ species) are presented in Table 1. Whitcomb, et al. (1963) reported observing the green lynx spider feeding on many species of Noctuidae, Geometridae and Phyralidae as well as Heliothis zea (Boddie), Alabama agrillacea (Hiibner) and Trichoplusia ni (Hiibner) in Arkansas cotton fields. In addition to those harmful insects, Whitcomb reported P. viridans feeding on A. mellifera , sphecid wasps, vespids of the genus Polistes and Dipternas, including syrphids and tachinids. Referring to the green lynx spider Weems and Whitcomb (1977) stated, “Judg- ing from their local abundance, the lynx spiders are among the major predators of insects occurring in the low shrubs and herbaceous vegetation.” They go on to state, however, “. . .their (P. viridans) usefulness in the control of insect pests is counteracted by their willingness to prey also upon beneficial insects.” Data collected in this investigation indicate that the green lynx spider is counter- productive as a predator of economically important insects since it takes beneficial insects as prey more often than it takes harmful insects. Eliminating the neutral grades (0), the ratio of beneficial to harmful prey taken by P. viridans in this study was 44:12. This ratio may change as further prey investigations of this type are conducted for the green lynx spider and by the biases of the specialists qualifying the prey as harmful or beneficial. Data in Table 1 indicate P. viridans to be a general insect feeder. The data lack the numbers required to show that the spider takes these insects in amounts sufficient to effect the overall prey populations. These data parallel those reported by Turner (1979). More extensive investigations, including data on the relative abundance of prey species, are needed to indicate that the green lynx spider is a possible biological control agent, or a counterproductive insect predator. RANDALL-GREEN LYNX SPIDER PREY RECORDS 21 Table l.-List and evaluation of prey of the green lynx spider, Peucetia viridans Hentz. ORDER FAMILY SPECIES No. Spms. -3 -2 -1 0 +1 +2 +3 Hymenoptera Ichneumonidae Ceratogaster ornata (Say) 2 * Anomalon sp. 1 * Sphecidae Arnmophila placida Smith Lins (Leptolarra) argintata 2 * (Palisot-Beauvais) 1 * Vespidae Vespula maculata (Linn.) 1 * Colletidae Colletes mitchelli Stephan 4 * Apidae Bombus impatiens (Cresson) 3 * Pompilidae Paracyphononyx fumerus (Lepeletier) 2 * Scoliidae Campsomersus plumipes (Fabr.)2 * Tiphiidae Myzinum sp. 1 * Myzinum prob. namea (Fabr.) 2 * Halictidae Halictus ligatus (Say) 5 * Chrysididae Omalus sp. 1 * Diptera Tachinidae Trichopoda pennipes (Fabr.) 2 * Bombyliidae Exoprosopa fasciata Macquart 2 * Dolichopodidae Condylostylus sp. 1 * Calliphoridae Cochliomyia macellaria (Fabr.) 1 * Phanenicia cuprina (Weid) 3 * Syrphidae Eristalis dimidiata Weidemann 8 * Sarcophagidae Sarcodexia innota (Walker) 1 * Hemiptera Pentatomidae Brochymena sp. 1 * Euschistus servus (Say) 3 * Phymatidae Phymata mystica Evans 2 * Reduviidae Zelus bilbobus Say 2 * Pselliopus cinctus (Fabr.) 1 * Corediae Leptoglossus phyllopus (L.) 3 * Rhopalidae Harmostes reflexulus (Say) 2 * Lepidoptera Noctuidae Genus ? 3 * Mocias latipes (Guenee) 2 * Coleoptera Scarabaeidae Anomala innuba (Fabr.) 1 * Cerambycidae Arhopalus nubilus (LeC.) 1 * Totals 66 3 2 7 10 19 2 23 % 4 3 10 15 29 3 35 Speculation on the potential utility of spiders as biological control agents has been great. Before the economic potential of spiders can be accurately determined quantitative and qualitative field prey data must be collected and evaluated. Qualified evaluation of economic importance, harmful or beneficial, or all prey taken by a spider must be included in any prey investigation. Without such data, conclusions on the usefulness of spiders as biological control agents, negative or positive, or their role in insect pest regulation will be misleading. ACKNOWLEDGMENTS I thank E. Grissell, F. Mead, H. Weems, D. Habeck and R. Woodruff for identifying and evaluating the prey specimens. 22 THE JOURNAL OF ARACHNOLOGY LITERATURE CITED Dondale, C. D. 1958. Note on the population densities of spiders (Araneae) in Nova Scotia apple orchards. Canadian Entomol., 90:11-113. Eason, R. and W. H. Whitcomb. 1965. Life history of the dotted wolf spider, Lycosa punctulata Hentz (Araneida :Lycosidae). Arkansas Acad. Sci. Proc., 19:11-20. Huffaker, C. B. and P. S. Messenger. 1976. Theory and Practice of Biological Control. Academic Press, New York, 778 pp. Peck, W. B. and W. H. Whitcomb. 1970. Studies on the biology of a spider Characanthium inclusum (Hentz). Univ. Arkansas Agr. Expt. Sta. Bull. 752, 1-76. Putman, W. L. 1967. Prevalance of spiders and importance as predators in Ontario peach orchards. Canadian Entomol., 99:160-170. Robinson, M. H. and B. Robinson. 1979. Prey caught by a sample population of the spider Argiope argentata (Araneae :Araneidae) in Panama: A year’s census data. Zool. J. Linn. Soc., 49:345-358. Turnbull, A. L. 1965. Effects of prey abundance on the development of the spider Agelenopsis potteri (Black well) (Araneae:Agelenidae). Canadian Entomol., 97:141-147. Turner, M. 1979. Diet and feeding phenology of the green lynx spider, Peucetia viridans (Hentz) (Araneae: Oxyopidae). J. Arachnol., 7:149-154. Weems, H. V. and W. H. Whitcomb. 1977. The green lynx spider, Peucetia viridans (Hentz) (Araneae: Oxyopidae). Florida Dept. Agr. Consumer Serv. Div. Plant Industry, Entomol. Cir. No. 181. Whitcomb, W. H., H. Exline and R. C. Hunter. 1963. Spiders of the Arkansas cotton field. Ann. Entomol. Soc. Amer., 56:65 3-660. Whitcomb, W. H., M. Hite and R. Eason. 1966. Life history of the green lynx spider, Peucetia viridans (Araneida: Oxyopidae). J. Kansas Entomol. Soc., 39(2):259-267. Whitcomb, W. H. and R. Eason. 1967. Life history and predatory importance of the striped lynx spider (Araneida: Oxyopidae). Arkansas Acad. Sci. Proc., 21:54-58. Whitcomb, W. H. 1973. Natural populations of entomophagous arthropods and their effects on the agroecosystem. Proc. Mississippi Symp. Biocontrol, Univ. Press Miss., 150-169. Yeargan, K. V. and W. R. Cothran. 1974. Population studies of Pardosa ramulosa (McCook) and other common spiders in alfalfa. Environ. Entomol., 3:989-993. Manuscript received October 1980, accepted January 1981. Bultman, T. L., G. W. Uetz and A. R. Brady. 1982. A comparison of cursorial spider communities along a succesional gradient. J. Arachnol., 10:23-33. A COMPARISON OF CURSORIAL SPIDER COMMUNITIES ALONG A SUCCESSIONAL GRADIENT Thomas L. Bultman and George W. Uetz Department of Biological Sciences University of Cincinnati Cincinnati, Ohio 45221 and Allen R. Brady Department of Biology Hope College Holland, Michigan 49423 ABSTRACT Wandering spiders from three communities representing points on a successional gradient (old field, oak and beech-maple forests) were sampled with pitfall traps and compared at the levels of species, family and guild. There was little similarity (species overlap) between communities. Species diversity was highest in the sub-climax forest and considerably lower in the mature beech-maple forest. This successional trend in species diversity is discussed in light of current hypotheses. Analysis of guild composition showed that, with succession, the relative abundance of wolf spiders decreased while that of vagrant web builders and crab spiders increased. Of the seven families sampled, only members of the Clubionidae occurred in fairly sizeable proportions in each community. Structure of the litter is discussed as a factor influencing cursorial spider abundance and distribution. INTRODUCTION There has been a sizeable amount of data of a descriptive nature collected on spider communities. Studies have revealed that specific plant associations harbor distinct spider faunas (Almquist 1973a, Chew 1961, Drew 1967,Duffey 1962, Elliot 1930,Kajak 1960, Muma 1973). Comparative studies have shown that spider community composition changes with vegetative succession. Changes in family and species composition with ecological succession have been reported by Gibson (1947), Dowdy (1950), Barnes (1953), Barnes and Barnes (1954) and Penniman (1975). Others (Berry 1967, Huhta 1971, Lowrie 1948) have also noted a general increase in species diversity through early and mid succession and a subsequent decrease in spider diversity in the climax com- munity. This trend in diversity has been suggested for community development in general 24 THE JOURNAL OF ARACHNOLOGY (Margalef 1968, Odum 1969), and has been reported by workers studying organisms other than spiders, such as birds (Johnston and Odum 1956, Kricher 1973) and plants (Pielou 1966, Whittaker 1969, 1975). Luczak (1959, 1963) and Duffey (1966, 1970) have suggested that physiognomy of plant communities is an important determinant of spider community composition, in that it influences microhabitats available to spiders. Others (Almquist 1970, 1973a, 1973b, Huhta 1971, Kuenzler 1958, Norgaard 1951, Vogel 1972) have demonstrated the impor- tance of microclimatic conditions in effecting observed spider distributions. Undoubt- edly, these two factors are closely tied; one might expect changes in microclimate to accompany changes in plant structure. Therefore, changes in plant structure during succession should result in community compositional changes in resident spider faunas. Guilds are ecological groupings of organisms which exploit a single or similar resources in a similar manner (Root 1967). Wandering, or cursorial spiders may be considered a “super-guild” or divided into several guilds depending upon their specific method of prey capture. Comparative studies of serai and climax communities which employ functional (rather than taxonomic) units of measure, such as guilds, may be useful because they deal with broad ecological roles common to most communities. This study examines the species composition, species diversity and guild structure of cursorial spiders in three communities along a successional gradient. MATERIALS AND METHODS Spiders were sampled from three isolated sites in western Michigan. An abandoned field (approximately 7 years old) located on the west side of Michigan State Route 40 near the City of Holland, contained the following dominant plants: Solidago spp. (golden rod), Daucus carota L. (Queen Anne’s lace), Carduus spp. (thistle), Vida spp. (vetch) and several sapling Acer negundo L. (boxelder). The site was rather heterogenous, and was bordered on the west by a small stream. Consequently there was an apparent gradient in humidity within the site. A second site, an edaphic sub-climax oak forest, was located 3 km NE of Fennville, near Allegan State Forest. It was characterized by Quercus velutina Lam. (black oak) and Quercus alba L. (white oak) in the canopy layer while Vaccinium vacillans Torr. (dryland blueberry) and Viburnum acerifolium L. (arrow-wood) were dominant shrubs. The herb layer was dominated by Car ex pensylvanica Lam. (sedge) and Pteridium aquilinum (L.) Kuhn, (bracken fern). The third site was located in the Hope College Biological Field Station, 12 km SW of Holland. It was a beech-maple climax forest, the typical end point of Lake Michigan dune succession (Clements 1936, Cowles 1899, Olson 1958). Dominant trees within the canopy were Fagus grandi flora Ehrh. (beech), Acer saccharum Marsh (sugar maple), and to a lesser degree, Prunus serotina Ehrh. (black cherry) and Quercus borealis Michx. (northern red oak). Viburnum sp. was the dominant shrub while Polygonatum biflorum (Walt.) Ell. (true Solomon’s seal) and Mitchella repens L. (Partridge -berry) dominated the herb layer. Spider communities in these sites were sampled with pitfall traps. Polypropylene cup traps (15 cm dia.) were fitted within a metal sleeve which was placed in the ground flush with the soil surface. A preservative (ethylene glycol) was placed in the bottom of the traps to a depth of about 5 cm. To prevent the accumulation of rain and leaves, the traps were covered with a square wooden roof, rasied 3 cm above the soil surface by four legs. Six pitfall traps were placed approximately 6 m apart along a transect within each study site. This appears to be the minimum number (of 15 cm dia. traps) necessary for accurate BULTMAN, UETZ AND BRADY-CURSORIAL SPIDER COMMUNITIES COMPARISON 25 sampling of cursorial spiders (see Uetz and Unzicker 1976). Traps were emptied weekly for 12 weeks from June to September 1977. Adult spiders were identified to species and, when possible, juveniles were separated and counted as morphospecies. These identifica- tions were used in community comparisons at the guild, family and species levels. The validity of pitfall trapping has been questioned because weather factors, differen- tial species activity and trap placement all influence results (Greenslade 1964, Turnbull 1973, Southwood 1966). Proponents (Breymeyer 1966, Gist and Crossley 1973) defend the method because it allows continuous sampling and is not limited to specific habitats. Recently it has been shown to be an adequate estimator of the number of species of cursorial spiders over a wide range of habitats (Uetz and Unzicker 1976). Although it does not give a true estimate of density, it does sample the number of cursorial spiders moving in an area for a given time (or the “active density” (Uetz 1977).) Spider species diversity in each community was calculated using the total information content index of Shannon (1948). The Shannon index (H') has been used in previous studies dealing with pitfall trapping of cursorial spiders (Jocque 1973, Uetz 1975,1976, 1979, Uetz et al. 1979) and appears to be the best available index for pitfall samples (see Pielou 1966). It takes the form: H' = s -2 Pi log* Pj where pj = proportion of total individuals in species i, and s = number of species. J' (H'/H'max), the estimate of the component of eveness, was also calculated. The Bray- Curtis similarity index, 2 i = 1 N, r Nj. + n2. i=i ■ * where N1; and N2j are the numbers of the ith species in communities 1 and 2, after log tranformation of the data (ln[X + 1]) (see Clifford and Stephenson 1975), was calcu- lated for spiders by community and by guild. RESULTS A total of 568 individuals was collected and included 41 species and 7 families of wandering spiders (Table 1). The old field samples contained the most individuals (243), representing 21 species and 5 families. Those from the oak forest contained 142 individ- uals and 21 species while the samples from the beech-maple forest contained 183 indi- viduals and 11 species. Samples from each forests contained representatives from 5 dif- ferent families. Calculated species diversity and evenness indices increase slightly from the old field to the oak forest and then decrease in the beech-maple forest (Table 1). Each community, as a whole and by guilds, is quite dissimilar from the others as indicated by values of the Bray-Curtis similarity index (Table 2). However, the two forests are more similar to each 26 THE JOURNAL OF ARACHNOLOGY Table 1.- Abundance and diversity of wandering spiders over the successional gradient. Old Field Oak Beech-Maple VAGRANT WEB BUILDERS Agelenidae Grcurina brevis (Emerton) C. pallida Keyserling C. robusta Simon Hahniidae Neoantistea magna (Keyserling) RUNNING SPIDERS Gnaphosidae Drassyllus aprilinus (Banks) D. depressus (Emerton) 6 Herpyllus ecclesiasticus Hentz Litophyllus temporarius Chamberlin Sergiolus decoratus Kaston 4 Zelotes hentzi Barrows Z. subtenaneus (C. L. Koch) 2 Clubionidae Agroeca sp. Castianeira cingulata (C. L. Koch) C. gertschi Kaston 4 C. variata Gertsch 5 Gubiona abbotii (L. Koch) 17 C. johnsoni Gertsch Micaria elizabethae Gertsch 2 M. pulicaria (Sundevall) 13 Phrurotimpus alarius (Hentz) P. borealis (Emerton) WOLF SPIDERS Lycosidae Lycosa frondicola Emerton L. gulosa Walckenaer L. modesta (Keyserling) Pardosa modica (Blackwall) 1 P. moesta Banks 75 P. saxatilis Emerton 29 Pirata minutus Emerton 19 Schizocosa avida Walckenaer 1 S. bilineata (Emerton) 36 S. ocreata (Hentz) S. crassipalpata Roewer 4 S. saltatrix (Hentz) Trabea aurantiaca (Emerton) 6 Trochosa terricola Thorell 7 CRAB SPIDERS Philodromidae Thanatus striatus C. L. Koch 2 3 3 2 1 2 80 2 1 1 1 8 28 15 3 3 1 19 24 1 33 2 24 1 9 2 2 1 BULTMAN, UETZ AND BRADY-CURSORIAL SPIDER COMMUNITIES COMPARISON 27 Table l.-(Cont.) Old Field Oak Beech-Maple Thomisidae Ozyptila conspurcata Thorell Xysticus elegans Keyserling 4 9 2 X. ferox (Hentz) 5 6 X. f rat emus Banks X. luctans (C. L. Koch) 1 5 31 N 243 142 183 S 21 21 11 H' (base 2) 3.380 3.575 2.306 J' 0.770 0.814 0.677 other than to the old field, and the oak forest is more similar to the old field than is the beech-maple forest. Some differences between communities are apparent in the analysis of guild composi- tion (Figure 1). The wolf spider guild declines sharply over the successional gradient. In addition, the vagrant web-building spiders are absent in the old field but dominate in the climax community. A steady increase in the relative abundance of crab spiders occurs over the successional sequence. In contrast, the running spiders show no distinct trends with succession. Family composition (Figure 2) tends to follow guild composition, but the sizable contributions of the Clubionidae and Hahniidae to the relative abundance of their guilds are apparent. Individuals of the Hahniidae were all of the species Neoantistea magna (Keyserling) (Table 1). Dominance-diversity curves (Whittaker 1975) for the three communities graphically show differences in community structure (Figure 3). The curves for the old field and oak forest approach lognormal distributions. In contrast, community structure of the beech- maple forest yields a curve approaching a geometric series. DISCUSSION Each of the plant communities is distinct in its cursorial spider species composition (Table 1). In fact, no species overlap occurs between the old field and the climax forest (Table 2). These results are in accord with those of previous researchers, who have found disparate communities when sampling dissimilar habitats along a successional gradient (Berry 1967, Huhta 1971, Lowrie 1948). These workers have also suggested that spider communities may show a pre-climax peak in species diversity. Odum (1969) proposed that the mechanism for a pre-climax peak in species diversity is the mixing of transitional species. In climax communities, diversity may decline some- what because transitional species, which cannot adequately compete with better adapted climax species, are forced out. Similarity values (Table 2) give some evidence in support of this mechanism. The oak forest contains some species which are common to the old field and some which are common to the beech-maple (and hence, contains some transi- tional species). Consequently, the oak forest is more similar to the old field and the beech-maple forest than the old field is to the beech-maple forest. Cannon (1965), 28 THE JOURNAL OF ARACHNOLOGY Table 2.-Bray-Curtis similarity values (after log transformation) calculated for spider communities in the three study sites. GUILDS O-F/OAK SITES OAK/B-M O-F/B-M Vagrant Web Builders .00 .44 .00 Running Spiders .07 .47 .00 Wolf Spiders .15 .22 .00 Crab Spiders .32 .55 .00 Total .14 .43 .00 studying forests and old fields in south central Ohio, has obtained similar results. A somewhat different mechanism has been put forth by Auclair and Goff (1971). In forests of the Great Lakes regions they found that some plant species within the climax com- munity are competitively superior (i.e., the shade tolerant beech and sugar maple) and become dominant. As a consequence, species diversity may decline considerably in older climax forests. Loucks (1970), studying similar forests, proposed that considerable declines in diversity are prevented by periodic burning. These perturbations act to return the forest to an earlier sere in which diversity is higher. The beech-maple forest exhibits a much lower value of cursorial spider species diver- sity than the other two communities (Table 1) despite a deep and well-developed litter layer. Species diversity of wandering spiders has been found to increase with increased litter depth in sub-climax oak-hickory and oak-tuliptree-maple forests (Uetz 1979) and over succession with increased litter depth (Huhta 1971). Our results fit this trend and 100- OOOOOOOOOOOI - o DO O lOOOOOOOOOOOOOOOOOC o0o0o°o0o0o0o%0o0o^^^ lOOOOOOOOOOOOOOOOOC ^o%%%%%0o0o0o0c0oVo% Crab spiders Vagrant web builders - Running spiders Wolf spiders O-F OAK B-M Fig. 1. -Guild composition of spider communities from the three study sites. O-F - Old Field and B-M = Beech-Maple. BULTMAN, UETZ AND BRADY-CURSORIAL SPIDER COMMUNITIES COMPARISON 29 suggest further that with extremely deep litter (i.e., like that of beech-maple forests) diversity declines. Dominance-diversity curves (Figure 3) graphically show the decrease in spider diversity (i.e., a steep slope) in the climax community. Similar results have been obtained in work on forest succession at Brookhaven, New York (Whittaker 1975). The dominance of the hahniid, Neoantistea magna , may be a major factor influencing the slope of the curve, in that it was the most abundant species trapped and occurred only in the beech-maple forest. The absence of fire in this preserve may explain the observed decline in spider diversity, although we have no historical data. In contrast, the high diversity of the old field may be explained in part from the fact that it has been undis- turbed for at least 7 years and has a diversity of plant species and structures. Nicholson and Monk (1974) have found old fields on the Georgia piedmont to double in plant species richness within the first decade of succession. Recently, spiders have been subdivided into increasingly finer guild systems (for example, see Post and Riechert 1977). Unfortunately, the present paucity of knowledge of foraging methods of some families of spiders makes the development of highly resolved guild systems difficult. The present system delineates four guilds, based upon gross differ- ences in foraging behavior within the cursorial spider community. Members of the wolf spider guild are “sit and wait” type predators which change sites frequently (Ford 1977). Crab spiders, while also being “sit and wait” type predators, are differentiated because of morphological differences (the first two pairs of legs are laterigrade rather than prograde). Because the foraging methods of the Clubionidae and Gnaphosidae are not well known, they have been put into a separate guild, the running spiders. They appear to be active pursuing predators, according to Gertsch (1979). The vagrant web builders are repre- sented by some members of the Agelenidae and Hahniidae (Uetz 1975), and while they 100 c5* w 75 a> o c <0 TJ C n 50 < a> > 2 * 25 CC iiiimiHtirf |~]] — Thomisidae n -Philodromidae -Hahniidae -Agelenidae -Clubionidae -Gnaphosidae -Lycosidae OF OAK BM Fig. 2. -Family composition of spider communities from the three study sites. 30 THE JOURNAL OF ARACHNOLOGY are known to spin small webs, they also leave their webs and wander through and over the litter while foraging. Guild composition in the three communities studied show some interesting trends (Figure 1). Members of the wolf spider guild dominate in the old field but then fall dramatically along the successional gradient. This scarcity of the wolf spider fauna in beech-maple forests has also been noted, although not to this degree, by Lowrie (1948). Members of the Lycosidae are typical field inhabitants and have been collected there in large numbers by several workers (Berry 1967, 1970, Doane and Dondale 1979, Peck 1966, Whitcomb et al. 1963). They seem best suited for locomotion in habitats where little litter accumulates. Uetz (1979) has found that augmenting natural litter depth results in a decrease in the relative dominance of wolf spiders while removal of litter increases their dominance. The development of a thick and intricate litter system during succession may prevent lycosids from dominating the cursorial spider community of climax forests. Increases in the relative abundance of the crab spiders and vagrant web builders are also noticeable. These spiders live within the litter and have previously been found to increase in dominance with the addition of litter (Uetz 1979). Family compositional changes (Figure 2) show that differences in vagrant web builders are, for the most part, due to the occurrence of hahniids in the climax community. In fact, differences in the relative abundance of both guilds are primarily determined by single species (Table 1): Neoantistea magna and Xysticus fraternus Banks. N. magna was the most commonly collected species in a study of beech forest in central Ohio (Penniman 1975) and appears to be a dominant typical of the climax forest. An increase in spider abundance has also been correlated with an increase in litter depth (Berry 1967, Hagstrum 1970, Lowrie 1948). Unfortunately no litter data are available from the study sites. However, it seems certain that the development and modification of the litter, which normally occurs with succession, is an important factor influencing the abundance of these spiders. Increased litter depth may enhance resource Fig. 3. -Dominance-diversity curves for spiders sampled from old field, oak and beech-maple communities. BULTMAN, UETZ AND BRADY-CURSORIAL SPIDER COMMUNITIES COMPARISON 31 partitioning by members of these guilds by increasing prey densities and/or microhabitat diversity, which may in turn allow a reduction of inter- and intraspecific competition and predation. While we have no quantitative litter data, this study does provide strong indirect evidence that cursorial spider abundance and distribution are closely correlated with litter development. ACKNOWLEDGMENTS This research was supported by National Science Foundation Grants 300 NSF DEB78- 03561 and NSF URP SP1 76-8361 2 to A. R. Brady and the Department of Biology, Hope College, respectively. We thank Hope College for providing the Biological Field Station for our use during this study. Special thanks to Kitty Uetz, who typed the manuscript. LITERATURE CITED Almquist, S. 1970. Thermal tolerances and preferences of some dune-living spiders. Oikos, 21:229-234. Almquist, S. 1973a. Spider associations in coastal sand dunes. Oikos, 24:444-457. Almquist, S. 1973b. Habitat selection by spiders on coastal sand dunes in Scandia, Sweden, Ent. Scand., 4:134-154. Auclair, A. W. and F. G. Goff. 1971. Diversity relations of upland forests in the western Great Lakes area. Amer. Natur., 105:499-528. Barnes, R. D. 1953. The ecological distribution of spiders in non-forested maritime communities at Beaufort, North Carolina. Ecol. Monogr., 23:315-337. Barnes, B. M. and R. D. Barnes. 1954. The ecology of the spiders of maritime drift lines. Ecology, 35:25-35. Berry, J. W. 1967. The distributional ecology of spiders in the old-field succession of the Piedmont region of North Carolina. Ph.D. Thesis, Duke University, Durham, N. C. Berry, J. W. 1970. Spiders of the North Carolina Piedmont old-field communities. J. Elisha Mitchell Sci. Soc., 86:97-105. Breymeyer, A. 1966. Relations between wandering spiders and other epigeic predatory Arthropoda. Ekol. Polsk, A. 14:18-71. Cannon, S. S. 1965. A comparison of the spider fauna of four different plant communities found in Neotoma, a small valley in south central Ohio. Ohio J. Sci., 65:97-110. Chew, R. M. 1961. Ecology of the spiders of a desert community. J. New York Entomol. Soc., 69:5-41. Clements, F. E. 1936. Nature and structure of the climax. J. Ecol., 24:252-284. Clifford, H. T. and W. Stephenson. 1975. An introduction to numerical classification. Academic Press, New York. 229 pp. Cowles, H. C. 1899. The ecological relations of the vegetation on the sand dunes of Lake Michigan. Bot. Gaz., 27:95-117, 167-202, 281-308, 361-391. Doane, J. F. and C. D. Dondale. 1979. Seasonal captures of spiders (Araneae) in a wheat field and its grassy borders in central Saskatchewan. Canadian Entomol., 111:439-445. Dowdy, W. W. 1950. A community study of an oak-hickory association with special references to invertebrates. Amer. Midi. Nat., 43:667-695. Drew, L. C. 1967. Spiders of beaver island. Pubis. Mich. St. Univ. Mus., Ser. 3; 153-208. Duffey, E. 1962. A population study of spiders in limestone grassland. Oikos, 13:15-34. Duffey. E. 1966. Spider ecology and habitat structure. Senckenb. Biol., 47:45-49. Duffey, E. 1970. Habitat selection by spiders in Saltmarsh in Gower. Nature in Wales, 12:15-23. Elliot, F. R. 1930. An ecological study of the spiders of the beech-maple forest. Ohio J. Sci., 30:1-22. Ford, M. J. 1977. Metabolic costs of the predation strategy of the spider Pardosa amentata (Clerk) (Lycosidae). Oecologia, 28:333-340. Gertsch, W. J. 1979. American spiders. 2nd Ed. Van Nostrand Reinhold Co., New York. 272 pp. 32 THE JOURNAL OF ARACHNOLOGY Gibson, W. W. 1947. An ecological study of the spiders of a river terrace forest in western Tennessee. OhioJ. Sci., 47:38-44. Gist, C. S. and D. A. Crossley. 1973. A method for quantifying pitfall trapping. Environ. Entomol., 2:951-952. Greenslade, P. J. M. 1964. Pitfall trapping as a means of studying Carabidae populations. J. Anim. Ecol., 33:301-310. Hagstrum, D. W. 1970. Ecological energetics of the spider Tarentula kochi. Ann. Entomol. Soc. Amer., 63:1297-1304. Huhta, V. 1971. Succession in the spider communities of the forest floor after clear-cutting and prescribed burning. Ann. Zool. Fenn., 8:483-542. Jocque, R. 1973. The spider fauna of adjacent woodland areas with different humus types. Biol. Jaarb., 41:153-179. Johnston, D. W. and E. P. Odum. 1956. Breeding bird populations in relation to plant succession on the Piedmont of Georgia. Ecology, 37:50-62. Kajak, H. 1960. Changes in the abundance of spiders in several meadows. Ekol. Pol. A, 8:199-228. Kricher, J. C. 1973. Summer bird species diversity in relation to secondary succession on the New Jersey Piedmont. Amer. Midi. Nat., 89:121-137. Kuenzler, E. J. 1958. Niche relations of three species of lycosid spiders. Ecology, 39:494-500. Loucks, O. L. 1970. Evolution of diversity, efficiency, and community stability. Amer. Zool., 10:17-25. Lowrie, D. C. 1948. The ecological succession of spiders of the Chicago area dunes. Ecology, 29:334-351. Luczak, J. 1959. The community of spiders of the ground flora of pine forests. Ekol. Pol. A, 14:233-244. Luczak, J. 1963. Differences in the structure of communities of web-building spiders in one type of environment (young pine forest). Ekol. Pol. A, 11:159-221. Margalef, D. R. 1968. Perspectives on ecological theory. Univ. of Chicago Press. Ill pp. Muma, M. H. 1973. Comparison of ground surface spiders in four central Florida ecosystems. Florida Entomol., 56:173-193. Nicholson, S. A. and C. D. Monk. 1974. Plant species diversity in old-field succession on the Georgia Piemont. Ecology, 55:1075-1085. Norgaard, E. 1951. On the ecology of two lycosid spiders ( Pirata piraticus and Lycosa pullata ) from Danish Sphagnum bog. Oikos, 3:1-21. Odum, E. P. 1969. The strategy of ecosystem development. Science, 164:262-270. Olson, J. S. 1958. Rates of succession and soil changes on southern Lake Michigan sand dunes. Bot. Gaz., 119:125-170. Peck, W. B. 1966. The population composition of a spider community in west central Missouri. Ameri. Midi. Nat., 76:151-168. Penniman, A. B. 1975. The ecology of the spiders of the old field succession in central Ohio. M.S. Thesis, Ohio St. Univ., Columbus, Ohio. Pielou, E. C. 1966. Species-diversity and pattern-diversity in the study of ecological succession. J. Theor. Biol., 10:370-383. Post, W. M. and S. E. Reichert. 1977. Initial investigation into the structure of spider communities. I. Competitive effects. J. Anim. Ecol., 46:729-249. Root, R. B. 1967. The niche exploitation pattern of the blue-gray gnatcatcher. Ecol. Monogr., 37:317-350. Shannon, C. E. 1948. A mathematical theory of communication. Bell Inst. Tech. J., 27:379-423; 623-656. Southwood, T. R. E. 1966. Ecological methods with particular reference to the study of insect populations. Methuen and Co., London. 391 pp. Turnbull, A. L. 1973. Ecology of the true spiders. Annu. Rev. Entomol., 18:305-348. Uetz, G. W. 1975. Temporal and spatial variation in species diversity of wandering spiders (Araneae) in deciduous forest litter. Environ. Entomol., 4:719-724. Uetz, G. W. 1976. Gradient analysis of spider communities in a stream-side forest. Oecologia, 22:373-385. Uetz, G. W. 1977. Coexistence in a guild of wandering spiders. J. Anim. Ecol., 46:5 31-542. Uetz, G. W. 1979. Influence of variation of litter habitats on spider communities. Oecologia, 40:29-42. BULTMAN, UETZ AND BRADY-CURSORIAL SPIDER COMMUNITIES COMPARISON 33 Uetz, G. W. and J. D. Unzicker. 1976. Pitfall trapping in ecological studies of wandering spiders. J. Arachnol., 3:101-111. Uetz, G. W., K. L. vanderLaan, G. F. Summers, P. A. K. Gibson, and L. L. Getz. 1979. The effects of flooding on floodplain arthropod distribution, abundance and community structure. Amer. Midi. Nat., 101:286-299. Vogel, B. R. 1972. Apparent niche sharing of two Pardosa species (Araneida: Lycosidae). Armadillo Papers, 7:1-13. Whitcomb, W. H., H. Exline and M. Hite. 1963. Comparison of spider populations of ground stratum in Arkansas pasture and adjacent cultivated field. Arkansas Acad. Sci. Proc., 17:34-39. Whittaker, R. H. 1969. Evolution of diversity in plant communities. Brookhaven Symp. Biol., 22:178-196. Whittaker, R. H. 1975. Communities and Ecosystems. 2nd ed. MacMillian, New York. 385 pp. Manuscript received August 1980, revised November 1980. ■ Francke, O. F. 1982. Are there any bothriurids (Arachnida, Scorpiones) in southern Africa? J. Arachnol., 10:35-40. ARE THERE ANY BOTHRIURIDS (ARACHNIDA, SCORPIONES) IN SOUTHERN AFRICA? Oscar F. Francke Department of Biological Sciences Texas Tech University Lubbock, Texas 79409 ABSTRACT As Gondwanaland fragmented due to plate tectonics, each of the southern continents carried with it a sample of the ancestral biota. Bothriurid scorpions are known from South America and Australia, and if this taxon was part of that ancestral biota then bothriurids would be predicted to occur in southern Africa as well. The genus Lisposoma Lawrence, currently placed in the Scorpionidae, lacks any demonstrable synapomorphies with other members of that family. The trichobothrial pattern and the structure of the tarsi represent synapomorphies between Lisposoma and the Bothriuridae. Lisposoma contains two species, both from Namibia, which represent the bothriurids in southern Africa. INTRODUCTION “The characters I have used are taken exclusively from the external structure. . . . The character that I believe to be new and, I hope, of considerable importance is the presence or absence of one of the spurs of the pair that is found upon the articular membrane connecting the foot or terminal segment of the legs with the segment that precedes it. . . . Of course it is hardly expected that this character, more than any other, will prove invariable', but it adds one more to the sum of characters upon which, as I have long suspected, the families or subfamilies of scorpions must be based.” Pocock 1893:303. (italics added) The scorpion family Bothriuridae is very interesting from a zoogeographic viewpoint. The two monobasic subfamilies Brachistosterninae and Vachonianinae are endemic to South America, while the Bothriurinae has eight genera in South America and one genus endemic to Australia and Tasmania (Maury 1973). Therefore, according the theory of vicariance biogeography (Platnick and Nelson 1978), either a bothriurid or their imme- diate sister group might be predicted to occur in southern Africa. A more specific predic- tion could be made if a cladogram expressing the phylogenetic relationships among bothriurids were available; however, Brachistosterninae and Vachonianinae were estab- lished on the basis of generic autapomorphies, and their phylogenetic relations to other bothriurid genera remain unknown. Nonetheless, the objective of this contribution is to test the hypothesis that a bothriurid scorpion does indeed occur in southern Africa. 36 THE JOURNAL OF ARACHNOLOGY THE SCORPION FAUNA OF SOUTHERN AFRICA The scorpions of southern Africa are relatively well known in comparison to those of other parts of the world (Hewitt 1918, 1925; Lamoral and Reynders 1975, Lamoral 1979, Lawrence 1955). Therefore, it is reasonable to assume that any bothriurids from that region are likely to have been collected and reported in the literature. Since there are no published records on bothriurids from southern Africa, it becomes quite possible that they have been classified under some other family to which they do not actually belong. There are only two scorpion families reported from Africa, Buthidae and Scor- pionidae. Buthids are so distinct, and so far removed phylogenetically from bothriurids that the likelihood of confusion is minimal. The scorpionids are represented in southern Africa by three subfamilies: Scorpioninae, Ischnurinae, and Lisposominae. The former two taxa contain several genera in both the Ethiopian and Oriental regions, and are well characterized. The Lisposominae, on the other hand, is monobasic and has been reported to have “affinities” with bothriurids (Lawrence 1928, Vachon 197'4). THE GENUS LISPOSOMA LAWRENCE, 1928 The genus Lisposoma Lawrence is endemic to Namibia. Its two recognized species, Lisposoma elegans Lawrence, 1928, and Lisposoma josehermana Lamoral, 1979, present a combination of external morphological features that made their taxonomic placement rather difficult. STERNUM: Lisposoma has a pentagonal sternum, a feature which under the present classification excludes it from the Buthidae (subtriangular sternum) and the Bothriuridae (sternum reduced to a narrow transverse sclerite). PEDAL SPURS: Lisposoma has prolateral pedal spurs, but lacks retrolateral pedal spurs, a feature which, if invariable, according to Pocock (1893; see opening quote above) places it within the Scorpionoidea (Scorpionidae and Diplocentridae). SUBACULEAR TUBERCLE: Lisposoma lacks a subaculear tubercle, which excludes it from the Diplocentridae. Thus, using a “process of elimination” approach to classification, Lisposoma turns out to be a scorpionid. However, it has several other features which exclude it from any of the recognized scorpionid subfamilies, which is why the Lisposominae was erected. How- ever, if we use a cladistic approach to classification, what is Lisposoma ? What is the phylogenetic importance of the characters used to place Lisposoma in the Scorpionidae? Are they primitive characters (i.e., plesiomorphies) or derived characters (i.e., apomorphies)? A pentagonal sternum is plesiomorphic among Recent scorpions, as indicated by ontogeny (both buthids and bothriurids have a pentagonal sternum as first, and sometimes second, instars). The loss of retrolateral pedal spurs is a derived character (by out-group comparison to buthids and eurypterids), but it has occurred independently at least three times: in Typhlochactas (Chactidae), in Vachonia, Thestylus and Phonio- cercus (Bothriuridae), and in the Scorpionoidea. Is the loss of retrolateral pedal spurs in Lisposoma the fourth independent occurrence of this transformation, or is it a synapo- morphy with scorpionoids, with some bothriurids, and/or with Typhlochactasl At pre- sent, the character transformation contains little useful information with respect to phylogeny at this level. Finally, the lack of a subaculear tubercle is considered plesio- morphic based on the fact that many more scorpions lack it than have it. Therefore, other characters need to be examined to understand the phylogenetic relations of Lisposoma. FRANCKE-ARE THERE ANY BOTHRIURIDS IN AFRICA? 37 OVARI UTERUS: The ovariuterus of Scorpionoidea bears numerous diverticula where embryonic development occurs; this is probably the most important apomorphic char- acter for the superfamily. I have examined adult females of both species of Lisposoma, and they lack ovariuteral diverticula. CHELICERAL DENTITION: Scorpionoids have been characterized as having a single subdistal tooth on the dorsal margin of the movable finger of the chelicera (Vachon 1963). Lisposoma , however, has two (as is the case in more chactids, vaejovids, and bothriurids). Two subdistal teeth appears to be a derived character (our group comparison with buthids and chaerilids), but there are some indications that it has appeared more than once among Recent scorpions, and thus will not be hypothesized to represent a synapomorphy among some scorpion taxa until further studies are made. TRICHOBOTHRIAL PATTERN: Vachon (1974) indicated that Lisposoma was very unusual among scorpionoids by having three trichobothria in line along the ventral articu- lation of the movable finger of the pedipalp, whereas all other scorpionoids with the same total number of trichobothria on the chela possess only two trichobothria in that position (some scorpionids have a very high number of trichobothria on the ventral aspect of the palm instead of the ‘usual’ 5-6, and in those, three can occur along the ventral hinge of the movable finger, although they are seldom in line). Vachon also pointed out that the pattern observed in Lisposoma is constant in all the bothriurid species that have the same total number of trichobothria as Lisposoma. This pattern appears to be unique to Lisposoma and bothriurids, and I hypothesize that it represents a synapomorphy between them. TARSI: In Lisposoma the tarsi are truncated distally, and the formula of ventral submedian spines is 2/2:3/2:3/3:3/3. Truncate tarsi with ventral submedian spines are unknown in scorpionids, but are characteristic of bothriurids (10 of 11 genera, the separation of which is based to a considerable extent on the spine formula: Brachis- tosternus is the exception, and their tarsal armature appears plesiomorphic). Further- more, I know of no chactids, iurids, or vaejovids with similar tarsi, and therefore hypoth- esize that tarsal structure represents a synapomorphy between Lisposoma and bothriurids (parsimony indicates that the more general character state is primitive). HEMISPERMATOPHORE: The hemispermatophores of Lisposoma spp. (Lamoral 1979) are unlike those of the scorpionid genera which have been studied [Scorpioninae: Scorpio (Vachon 1952a), Opisthophthalmus (Lamoral 1979, Francke, pers. obs.), Hetero- metrus (Francke, pers. obs.), and Pandinus (Vachon 1952b). Ischnurinae: Hadogenes (Francke, pers. obs.), Liochelis (Koch 1977), and Opisthacanthus (Francke, pers. obs.). Urodacinae: Urodacus (Koch 1977, Francke, pers. obs.)] with respect to both pedicel and capsule structure. In the structure of the pedicel and capsule of the hemispermatophore Lisposoma is very similar to bothriurids (Maury 1980); comparison of illustrations of the hemisperma- tophore of Lisposoma spp. (Lamoral 1979) with that of Timogenes sp. (Maury and San Martin 1973) reveals very interesting similarities (homologies ?) in the pedicel and cap- sule, whereas none could be found between Lisposoma and scorpionids. Bothriurid hemispermatophores are characterized by a prominent crest or ridge on the distoexternal aspect of the lamina (Maury 1980), a hypothesized autapomorphy for the family. Lamoral (1979) illustrated only the internal and dorsal views of the hemispermatophore of Lisposoma, and it is thus impossible to determine at this time whether a crest is present or not. 38 THE JOURNAL OF ARACHNOLOGY CONCLUSIONS I have been unable to find any synapomorphies between Lisposoma and other genera of Scorpionidae. The Scorpionoidea (Scorpionidae + Diplocentridae) are characterized by an ovariuterus with numerous diverticula, Lisposoma lacks this derived character state and is thus excluded from this superfamily. The loss of retrolateral pedal spurs has occurred independently at least three times among Recent scorpions, and this character provides no indication of the phylogenetic relations of Lisposoma. The trichobothrial pattern and the structure of the tarsi, however, represent synapomorphies between Lisposoma and the Bothriuridae. Therefore, I hypothesize that Lisposoma is indeed a bothriurid and not a scorpionid. Examination of the distoexternal aspect of the lamina of the hemispermatophore of Lisposoma for the presence of a crest could provide an inde- pendent test of the hypothesis formulated above. The only taxonomic change proposed here is the transfer of Lisposoma from the Scorpionidae to the Bothriuridae. The status of the Lisposominae, as well as the three recognized subfamilies of Bothriuridae, remains uncertain pending the construction of a cladogram expressing phylogenetic relationships within the family. Finally, the stated objective of this contribution was to test a hypothesis based on zoogeographic considerations: that a bothriurid scorpion should occur in southern Africa. A cladistic analysis of the phylogenetic relations of Lisposoma indicates that this is the African bothriurid sought. ACKNOWLEDGMENTS I am thankful to the late Dr. Erik N. K.-Waering for the opportunity to examine an adult female of Lisposoma elegans. My gratitude also goes to Mr. Alexis Harington, University of Witwatersrand, Johannesburg, for sending an adult female and an immature of Lisposoma josehermana for study. Finally, Dr. Emilio A. Maury, Museo Argentino de Ciencias Naturales, Buenos Aires, and Dr. Norman I. Platnick, American Museum of Natural History, New York, made valuable comments on the manuscript. Partial support was received from the Institute for Museum Research, Texas Tech Univer- sity. LITERATURE CITED Hewitt, J. 1918. A survey of the scorpion fauna of South Africa. Trans. Roy. Soc. South Africa, 6:89-192. Hewitt, J. 1925. Facts and theories on the distribution of scorpions in South Africa. Trans. Roy. Soc. South Africa, 12:249-276. Koch, L. E. 1977. The taxonomy, geographic distribution and evolutionary radiation of Australo- Papuan scorpions. Rec. West. Australian Mus., 5 (2): 8 3-367. Lamoral, B. H. 1979. The scorpions of Namibia. Ann. Natal Mus., 23(3):497-784. Lamoral, B. H. and S. C. Reynders. 1975. A catalogue of the scorpions described from the Ethiopian faunal region up to December 1973. Ann. Natal Mus., 22(2):489-576. Lawrence, R. F. 1928. Contributions to a knowledge of the fauna of South West Africa. VII. Arach- nida (Part 2). Ann. South African Mus., 25 (2): 217-31 2. Lawrence, R. F. 1955. Solifugae, Scorpions and Pedipalpi, with checklists and keys to the South African families, genera and species. Results of the Lund Univ. Exped. in 1950-1951. South African Animal Life, 1:152-262. FRANCKE-ARE THERE ANY BOTHRIURIDS IN AFRICA? 39 Maury, E. A. 1973. Essai d’une classification des sous-familles de scorpions Bothriuridae. 5th Int. Cong. Arachnol., Brno 1971, pp. 29-36. Maury, E. A. 1980. Usefulness of the hemispermatophore in the systematics of the scorpion family Bothriuridae. 8th Int. Cong. Arachnol., Vienna 1980, pp. 335-339. Maury, E. A. and P. R. San Martin. 1973. Revaluation del genero Timogenes Simon 1880. Physis, sect. C, 32(84):129-140. Platnick, N. I. and G. Nelson. 1978. A method of analysis for historical biogeography. Syst. Zool., 27:1-16. Pocock, R. I. 1893. Notes on the classification of scorpions, followed by some observations on synonymy, with descriptions of new genera and species. Ann. Mag. Nat. Hist., ser. VI, 12:303-330. Vachon, M. 1952a. Etudes sur les scorpions. Publ. Inst. Pasteur d’Algerie, Alger, pp. 1-482. Vachon, M. 1952b. Scorpions. Mission M. Lamotte en Guinee (1942). Mem. Inst. Fran§ais Afrique Noire, 19:9-15. Vachon, M. 1963. De Futilite, en systematique, d’une nomenclature des dents des cheliceres chez les scorpions. Bull. Mus. Nat. Hist. Nat., Paris, 2 ser., 35(2): 161-166. Vachon, M. 1974. Etudes des caracteres utilises pour classer les families et les genres de scorpions (Arachnides) 1. La trichobothriotaxie en Arachnologie. Sigles trichobothriaux et types de tricho- bothrioraxie chez les scorpions. Bull. Mus. Nat. Hist. Nat., Paris, 3 ser., No. 140, Zool. 104:857-958. Manuscript received July 1980, revised October 1 980. . Muchmore, W. B. and E. Hentschel. 1982. Epichernes aztecus, a new genus and species of pseudo- scorpion from Mexico (Pseudoscorpionida, Chernetidae). J. Arachnol., 10:41-45. EPICHERNES AZTECUS, A NEW GENUS AND SPECIES OF PSEUDOSCORPION FROM MEXICO (PSEUDOSCORPIONIDA, CHERNETIDAE) William B. Muchmore Department of Biology, University of Rochester Rochester, New York 14627 and Edna Hentschel Laboratorio de Acarologia, Facultad de Ciencias Universidad Nacional Autonoma de Mexico Mexico 20, D.F., Mexico ABSTRACT A new genus, Epichernes Muchmore, and species, E. aztecus Hentschel, are described from the Distrito Federal, Mexico. All specimens have been found on the bodies of the volcano mouse, Neo- tomodon alstoni alstoni Merriam. INTRODUCTION In a project designed by M. en C. Comelio Sanchez H. (Instituto de Biologia, U.N.A. M.) to study the ecology of a rodent community at El Ajusco, D.F., Mexico, it was found that many pseudoscorpions occurred on the volcano mouse, Neotomodon alstoni. One of us (E. H.) joined the project to study the pseudoscorpions and has reported on their ecology (Hentschel 1979). All the pseudoscorpions belong to a single species, at first thought to belong to the genus Dinocheirus Chamberlin. Further study, however, showed that they represent a new genus and species, which are described below. Epichernes Muchmore, new genus Type species. -Epichernes aztecus Hentschel, new species. Etymology. -Epichernes, masculine in gender, is from the Greek epi, above or near to, and Chernes, a genus of pseudoscorpions, and signifies a similarity to but difference from the genus Chernes. Diagnosis.— A genus of the family Chernetidae Chamberlin. Of moderately large size; generally heavily sclerotized, therefore, dark in color, with palps and carapace reddish to 42 THE JOURNAL OF ARACHNOLOGY dark brown. Vestitural setae sparsely denticulate or acuminate. Carapace with 2 deep transverse furrows; surface granulate; no eyes evident; with 150-200 setae. Tergites and sternites divided, surfaces scaly; pleural membranes longitudinally rugose and papillose; middle tergites and sternites with 20-30 marginal setae; 11th tergite with 4 and 11th sternite with 2 long, tactile setae; setae of spiracular plates acuminate, those of anal plates terminally denticulate. Cheliceral hand with 5 setae, b, sb and es denticulate; flagellum of 4 denticulate setae, the 2 basal ones short; galea of female moderate in size and branched, that of male small and denticulate. Palp robust, that of male slightly heavier than that of female; tibia with a prominent dorsomedial swelling; surfaces granulate; setae denticulate. Trichobothrium st on movable finger much nearer to t than to sb; on fixed finger ist at same level as, or proximad of, est; ib at about same level as est. Venom apparatus well developed in movable finger, much reduced in fixed finger; each finger well provided with both external and internal accessory teeth. Legs moderately slender; tarsus of leg IV with a prominent, acuminate tactile seta distad of middle. Anterior genital operculum of male with 4 large setae medially, flanked by many shorter ones; anterior operculum of female with a compact ^-shaped group of about 20 setae; spermathecae of female in form of 2 rather short, diverging tubes. The type species, at least, is found, apparently phoretic, on the bodies of small rodents of the genus Neotomodon. Remarks.— Superficially Epichernes appears most similar to Dinocheirus Chamberlin (see Muchmore 1974a), Epactiochernes Muchmore (1974b), and Mexachernes Hoff (1947). From the last it can be distinguished by setae b and sb on the cheliceral hand, both denticulate in Epichernes , both acuminate in Mexachernes. Epichernes differs from Dinocheirus in the nature of the spermathecae of the female, which are rather short, thick tubes in the former, but long, thin tubules with expanded ends in the latter. The sperma- thecae of Epichernes and Epactiochernes are somewhat alike, but Epichernes is much larger in body size and has many more setae on the carapace and abdomen than Epactiochernes. Because of the similarity between Epichernes and Dinocheirus , some of the Middle and South American species presently assigned to Dinocheirus may actually belong to the new genus. Epichernes aztecus Hentschel, new species Material examined.— The holotype male and 22 paratypes (9 males, 10 females, and 3 tritonymphs) were mounted on slides; there are also about 750 paratypes in alcohol. All specimens were collected from the hair of volcano mice ( Neotomodon alstoni alstoni Merriam) in a pine wood, elevation 2850 m, at El Ajusco, south of Mexico City, D.F., Mexico, from March 1978 to April 1979 (E. Hentschel). Types are deposited in the Arachnid Collection of the Acarology Laboratory (Facultad de Ciencias, U.N.A.M.). Description of male (based on the 10 mounted specimens).— Carapace and palps heav- ily sclerotized, dark brown to red; tergites light brown and the rest of the body paler. Carapace a little longer than broad, with 2 distinct transverse furrows and without eyes; surface heavily granulate, with more than 200 terminally dentate, vestitural setae, of which 7 are at anterior margin and about 20 near posterior margin. Tergites 1-10 and sternites 4-10 divided; surfaces of tergites granulate, of sternites almost smooth; inter- scutal and pleural membranes strongly papillose; most dorsal setae terminally denticulate, ventral setae acuminate. Tergal chaetotaxy of holotype 23:23:22:23:23:27:27:24:25:20:T8T:2; others varied. Sternal chaetotaxy of holotype MUCHMORE AND HENTSCHEE-EPICHERNES AZTECUS, NEW GENUS AND SPECIES 43 28:(2)(2-2)/(40)(2):(l)25(l):30:30:31:30:27:25:(T-T)/(T-ll-T):2; others varied, often with fewer setae on the posterior genital operculum. Setae on anterior genital operculum range 24-30 (Fig. 6); posterior operculum with 2 sets of 2 small setae beneath anterior margin and 30-40 setae scattered on face and posterior margin, those on margin longer than those on face; setae on stigmal plates acuminate and those on anal plates denticulate. Internal genitalia of the usual chernetid type, well sclerotized and distinct. Chelicera nearly 0.4 as long as carapace; hand with 5 setae, b and sb terminally denticulate, es acuminate or denticulate and shorter than b; flagellum of 4 setae, 2 large distal ones, denticulate along margins, and 2 short proximal ones, denticulate sub- terminally; galea short, with 0-5 very small subterminal denticulations. Palps rather robust, with chelal hands usually deeper than broad and tibia with a protuberance on the medial side (Figs. 1 and 3); palpal femur 2.0-2.29, tibia 1.73-2.04, Figs. 1-8 .—Epichernes aztecus, new species: 1, dorsal view of right palp of holotype male; 2, dorsal view of right palp of female; 3, lateral view of right chela of holotype male; 4, lateral view of right chela of female; 5, lateral view of leg IV; 6, genital opercula of holotype male; 7, genital opercula of female; 8, spermathecae of female. 44 THE JOURNAL OF ARACHNOLOGY and chela (without pedicel) 1.88-2.1 times longer than broad; hand (without pedicel) 0.94-1.1 times as long as deep; movable finger 1.0-1.16 times as long as hand. Surfaces more or less granulate, except chelal fingers; most setae terminally broadened and dentic- ulate. Trichobothria as indicated in Fig. 3. Fixed finger with 37-44 contiguous, cusped marginal teeth, and 9-12 external and 12-15 internal accessory teeth; movable finger with 40-44 similar marginal teeth and 9-11 external and 9-12 internal accessory teeth; venom apparatus well developed only in movable finger, with nodus ramosus about midway between trichobothria t and st; terminal tooth of fixed finger reduced. Legs rather slender; leg IV (Fig. 5) with entire femur 2.62-3.64 times as long as deep; tactile seta on tarsus rather long and erect, just distad of middle of tarsus. Description of female (based on the 10 mounted specimens).— Similar to male in most respects, but larger and with palpal chela slightly less robust. Genital opercula as shown in Fig. 7; anterior operculum with a compact fLshaped group of 19-24 setae on face, posterior operculum with 14-19 short setae along posterior margin. Spermathecae very delicate (Fig. 8) and often difficult to make out or lost entirely. Cheliceral galea much better developed than in male, with 5 or 6 prominent rami. Palps much as in male, with exception of chelal hand (Figs. 2 and 4); femur 1.95-2.15, tibia 1.89-2.13, and chela (without pedicel) 2.23-2.57 times as long as broad; hand (without pedicel) 1.0-1.08 times as long as deep; movable finger 1.16-1.27 times as long as hand. Fixed finger with 8-10 external and 10-12 internal and movable finger with 9-12 external and 5-8 internal accessory teeth. Tritonymph (based on 3 mounted paratypes).-Similar to adults, but smaller, paler, and with reduced number of setae on some structures. Carapace with about 150 setae. Tergal chaetotaxy of one paratype 16: 16: 15: 16: 16: 18:20: 16: 18: 16:T8T:2; sternal chaetotaxy 4: (2)6(2):(1)1 2(1): 16:20: 18:20: 17: 17:(T-T/(T-8-T):2. Cheliceral galea a small, simple elevation or elongate and with prominent rami. Trichobothria isb and sb absent from fixed and movable chelal fingers respectively, as is usual. Fixed finger with 32 and movable finger with 33 marginal teeth. Measurements (mm).— Male (figures given first for holotype, followed in parentheses by ranges for the 9 paratypes): Body length 3.42(2.42-3.72). Carapace length 1.04(0.98-1.15). Chelicera 0.40(0.36-0.42) by 0.20(0.17-0.23). Palpal trochanter 0.61(0.49-0.61) by 0.37(0.33-0.39); femur 0.88(0.78-1.03) by 0.41(0.33-0.50); chela (without pedicel) 1.50(1.37-1.73) by 0.68(0.68-0.92); hand (without pedicel) 0.81(0.65-0.90) by 0.77(0.67-0.94); pedicel about 0.15 long; movable finger 0.82(0.76-0.96) long. Leg IV: entire femur 0.91(0.78-0.95) by 0.25(0.24-0.29); tibia 0.80(0.72-0.92) by 0.18(0.15-0.21); tarsus 0.55(0.45-0.62) by 0.09(0.09-0.13). Female (ranges for the 10 paratypes): Body length 3.35-4.61. Carapace length 1.06-1.23. Chelicera 0.38-0.47 by 0.16-0.24. Palpal trochanter 0.47-0.54 by 0.35-0.40; femur 0.84-0.99 by 0.40-0.49; tibia 0.84-0.96 by 0.41-0.47; chela (without pedicel) 1.41-1.69 by 0.59-0.70; hand (without pedicel) 0.67-0.76 by 0.65-0.73; pedicel 0.13-0.14; movable finger 0.81-0.92 long. Leg IV: entire femur 0.84-1.02 by 0.26-0.29; tibia 0.78-0.93 by 0.17-0.20; tarsus 0.53-0.64 by 0.13-0.14. Tritonymph (ranges for the 3 mounted specimens): Body length 2.23-2.60. Carapace length 0.75-0.85. Chelicera 0.26-0.27 by 0.14-0.15. Palpal trochanter 0.33-0.37 by 0.15-0.24; femur 0.57-0.58 by 0.28-0.32; tibia 0.55-0.57 by 0.27-0.28; chela (without pedicel) 0.88-0.96 by 0.39-0.42; hand (without pedicel) 0.44-0.48 by 0.39-0.42; pedicel 0.08; movable finger 0.51-0.54 long. Leg IV: entire femur 0.62-0.65 by 0.24-0.26; tibia 0.52 by 0.14-0.17; tarsus 0.32-0.33 by 0.1 1-0.13. MUCHMORE AND HENTSCHEL- EP1CHERNES AZTECUS, NEW GENUS AND SPECIES 45 Etymology.— The species is named aztecus in reference to its occurrence in the region of Mexico where the Aztec culture was centered. Remarks.— The variation in measurements and proportions of Epichernes aztecus is considerable; the specimens chosen for description included the widest possible range of sizes. There is sexual dimorphism in the species in that the females are usually larger than the males and their palpal chelae are less robust. All of the 766 specimens collected were combed from the fur of live-trapped volcano mice (Neotomodon alstoni alstoni). Other rodents in the same vicinity carried no pseudo- scorpions. The pseudoscorpions appear to feed mainly upon the ectocommensal mites which also occur on the volcano mice in large numbers. The relations among the pseudo- scorpions, the mites, and the rodent host are discussed by Hentschel (1979). There is no obvious modification of E. aztecus for life on the body of the host mammal, such as that shown by Chiridiochernes platypalpus Muchmore (1972) where the palps are long and flattened, thus facilitating movement among hairs; E. aztecus has heavy palps with globose hands. However, E. aztecus may be protected against scratching by the host by its well sclerotized and hardened cuticle. ACKNOWLEDGMENTS E. H.— I am very grateful to M. en C. Cornelio Sanchez H. who kindly allowed me to work with his group, to all the people who contributed to the field collections, and to Dr. Ana Hoffmann for her help and advice. W. B. M.— This work was supported in part by the Office of Naval Research under Contract N00014-76-C-0001 with the Center for Naval Analyses of the University of Rochester. LITERATURE CITED Hentschel, E. 1979. Biologia del pseudoscorpion Dinocheirus sp. asociado a Neotomodon alstoni (Mammalia Rodentia). Thesis, Universidad Nacional Autonoma de Mexico, Mexico, D.F., 79 pp. Hoff C. C. 1947. The species of the pseudo scorpion genus Chelanops described by Banks. Bull. Mus. Comp. Zool., 98:473-550. Muchmore, W. B. 1972. A remarkable pseudo scorpion from the hair of a rat (Pseudoscorpionida, Chernetidae). Proc. Biol. Soc. Washington, 85:427-432. Muchmore, W. B. 1974a. Clarification of the genera Hesperochernes and Dinocheirus (Pseudo- scorpionida, Chernetidae). J. Arachnol., 2:25-36. Muchmore, W. B. 1974b. Pseudoscorpions from Florida. 3. Epactiochernes, a new genus based upon Chelanops tumidus Banks (Chernetidae). Florida Entomol., 57 :397-407. Manuscript received November 1980, revised February 1981. Richman, D. B. 1982. Epigamic display in jumping spiders (Araneae, Salticidae) and its use in system- atics. J. Arachnol., 10:47-67. EPIGAMIC DISPLAY IN JUMPING SPIDERS (ARANEAE, SALTICIDAE) AND ITS USE IN SYSTEMATICS David B. Richman Department of Entomology and Nematology University of Florida Gainesville, Florida 32611 ABSTRACT A study was made of visual epigamic displays in representative species of North American jumping spiders, with a special emphasis on determining the value of using such displays in systematics. Several presumed homologies in courtship behavior were described which largely support the most recent subfamilial arrangement on morphology. Courtship was found to be of more use in some generic and specific level problems, especially with closely related sympatric species. Agonistic displays were found to be of less use in systematics than courtship, because of the similar ways in which males engage in combat display, even in distantly related genera. INTRODUCTION The Salticidae (jumping spiders) is the largest of all the spider families, with approximately 3200 described species (Proszyriski 1971). These spiders are diurnal vagabond hunters, which visually orient toward prey. They capture prey by jumping on it. Salticids have a characteristic “squared-off’ prosoma (or cephalothorax) and are easily recognized by their enlarged anterior median eyes. The family is most diverse in the tropics, although a large fauna occurs in temperate areas. The species exhibit a great variety of form and coloration that has both intrigued and confused the systematists who have tried to work out a reasonable classification within the family. The males are often colorfully ornamented and thus have quite a different appearance than the females, which are usually more cryptic. This male ornamentation appears to function primarily in visual courtship (Peckham and Peckham 1889, 1890, Crane 1949). Platnick (1971) pointed out that courtship (defined here as intraspecific male-female interactions, which precede and are preparatory to successful mating) is probably the most useful of behaviors for species level systematics because elements of it serve as isolating mechanisms. In the case of jumping spiders, visual courtship, because of the complexity of the displays, is probably more useful than the non-visual (cohabitation) courtship which has been observed in some species, especially of Phidippus (Jackson 1976, 1977, 1978). Certain aspects of visual courtship such as the basic patterns of male movements or postures may also be more conservative than some other behaviors, such as cryptic or mimetic behaviors which are more related to environmental change, and may show similarities between closely related groups. Crane (1949) proposed a graded behavioral classification consisting of “runners,” “intermediates” and “hoppers.” Crane characterized each of these groups by their normal means of locomotion, courtship and 48 THE JOURNAL OF ARACHNOLOGY agonistic display (defined here as intraspecific male-male interactions, including those which are similar to courtship and those which involve fighting). This system was not meant to be a natural classification, but a series of evolutionary grades, through which several different genetic lines had passed. As Crane pointed out, under this system many displays may reflect parallelism among the subfamilies. Also the means of locomotion, which is used as a major criterion for the classification, is itself a very plastic adaptation to a specific environment. Crane (1949, text fig. 2) also proposed some courthsip characteristics (abdomen twisting, etc.) which might help define subfamilies. It is my contention that such visual courtship behavior can, with caution, be useful for the systematics of taxa ranging from the species level to the subfamilial level. Agonistic displays are less useful because they may not occur in all species and they contain fewer species-specific elements than does courtship. During the last eight years I have gathered observations and films of North American salticid courtships and I am now able to present preliminary findings on the application of visual courtship behavior to some systematic problems in the family Salticidae. MATERIALS AND METHODS Adult and immature salticids of 48 North American species are obtained by field collecting in southern Arizona [Metacyrba californica (Peckham and Peckham), Sarinda cutleri (Richman), Metaphidippus manni (Peckham and Peckham), M. vitis (Cockerell), Sassacus papenhoei Peckham and Peckham, Pellenes arizonensis Banks, P. clypeatus (Banks), P. cf, coecatus (Hentz), P. hallani Richman, P. hirsutus (Peckham and Peckham) and P. tarsalis Banks] , Ontario [Pellenes calcaratus Banks and P. viridipes (Hentz)] , Georgia [Pellenes coecatus (Hentz) and one male of Metacyrba undata (DeGeer)] and Florida (the rest of the species). These were maintained in the laboratory in plastic vials and fed fruit flies ( Drosophila melanogaster Meig.) or cabbage looper larvae [Trichoplusia ni (Hiibner)] . Immature specimens were reared to maturity. Apparently healthy individuals of both sexes were placed together in arenas of plexiglass or plastic (including petri dishes) and filmed or observed during their courtship or agonistic display. Most of the courtships were recorded on super-8 color film (high speed Ektachrome®) using Bolex® 280 macrozoom or Beauliue® 4008 movie cameras. In most cases visual courtships were filmed under artificial light with a film speed of 18 fps (normal speed), but one film was made in natural light and another was shot at twice normal speed. Analysis of the courtships was conducted through a frame-by-frame inspection of each film. Events were timed based on the film speed and ethograms of selected courtships were constructed. Virgin female salticids were, in general, more receptive to males than previously mated females, but with some species [Marpissa pikei (Peckham and Peckham), Sarinda cutleri (Richman), Hentzia mitrata (Hentz)] only mated females were available. Females of these and of Pellenes brunneus Peckham and Peckham (females of which were strongly resistant to mating in the laboratory) were chilled to allow the male to go through the entire display without interruption. In most other cases virgin females were used at room temperature (20-25°C). Generally, the male was introduced to the arena first and the female dropped down on her dragline in front of the male. Experiments were conducted with three species, one each of Crane’s behavioral groups, to determine how important pheromones might be in courtship. Filter paper was placed in petri dishes containing a female spider and left for 24 hrs. This was then cut in half and RICHMAN-EPIGAMIC DISPLAY IN JUMPING SPIDERS 49 placed with a clean half in a clean petri dish. A male of the same species was introduced and a filming speed of 2 fps was used to record the males’ actions for three minutes. The last two minutes of these films were used to analyze the male reaction and time spent on each half circle of paper. The technique is a modification of a method used by Hegdekar and Dondale (1969) for lycosid spiders. RESULTS AND DISCUSSION A summary of the courtships of 48 species of North American salticids is presented in this section. Some representative courtships are depicted in ethograms (Appendix Figures 1-13). Notes on agonistic displays are incorporated in the text, but as these are not as useful as visual courtship in systematics they will not be discussed in as much detail. The species of salticids are arranged in the subfamilies in which they seem to fit, based both on current morphological studies (Proszyriski 1976, Hill 1979) and courtship behavior. The subfamilies proposed by Petrunkevitch (1928, 1939) and utilized by Crane (1949) in her Table II have proven to be an inaccurate reflection of the phylogeny of the various genera in the family. Crane’s behavioral classification (R = “runner”, I = “intermediate”, H = “hopper”) as it would apply to each species is presented for comparison. Mountings are recorded in the text only for unchilled virgin females. Subfamily Lyssomaninae: Lyssomanes viridis (Walckenaer) (R). Early Movements: stationary for 5 sec at a time, then approached directly, repeat. Later Movements: as in early. Retinae of AME: moved back and forth. Palpi: held 90° to substrate (straight up) stationary. First Legs: extended laterally, tarsi-metatarsi jerked alternately at intervals, prior to forward motion. Prosoma: raised off substrate. Opisthosoma: bent downward. Number of Observations: 2. Number of Males: 2. Number of Mountings: 0. Sexual Dimorphism: male with elongated and enlarged chelicerae; front legs longer than in female. Crane (1949) noted the retinal movements of males of Lyssomanes bradyspilis Crane during courtship. These eye movements were also noted in L. viridis (Walck.), but they have been also observed in relation to prey capture in other salticids (Bristowe 1941, Land 1969). These may thus be simply scanning movements by the retinae of the male, which are accelerated because of the heightened stimulus of the female presence. The unusually transparent prosoma of Lyssomanes makes such retinal movements more noticeable then they would be in other salticids. The species of Lyssomanes , because of their unique morphology (both in eye position and epigynal and palpal structure) and their distinctive courtship, warrant at least a separate subfamily. Subfamily Marpissinae: Marpissa bina (Hentz) (R). Early Movements: direct toward female. Later Movements: zigzag. Palpi: lowered, stationary. First Legs: raised. Prosoma: low to substrate. Opisthosoma: raised. Number of Observations: 1. Number of Males: 1. Number of Mountings: 0. Sexual Dimorphism: sexes very similar; males usually darker, females with central light band on opisthosoma. M. pikei (Peckham and Peckham) (R). Early Movements: direct. Later Movements: direct. Palpi: lowered, stationary. First Legs: raised. Prosoma: low or occasionally slightly raised off substrate. Opisthosoma: raised. Number of Observations: 2. Number of Males: 1. Number of Mountings: 0. Sexual Dimorphism: sexes differ primarily in pattern, males being darker with central dark band on opisthosoma. 50 THE JOURNAL OF ARACHNOLOGY M. sulcosa Barnes (R). Early Movements: zigzag. Later Movements: zigzag. Palpi: stationary. First Legs: raised. Prosoma: raised. Opisthosoma: raised, twisted to right or left. Number of Observations: 6. Number of Males: 1. Number of Mountings: 0. Sexual Dimorphism: sexes very similar. Maevia inclemens (Walckenaer) (H). Early Movements: zigzag. Later Movements: direct. Palpi: mostly stationary. First Legs: raised, moved back and forth. Prosoma: raised. Opisthosoma: straight. Number of Observations: 5. Number of Males: 1. Number of Mountings: 0. Sexual Dimorphism: “normal” phase male similar to female, except for having yellow hairs on palpi and stripes and spots on femora of legs; black phase male differs from female in having black body and three tufts above anterior eyes. This subfamily, as characterized by Proszyriski (1976), bears little resemblance to the subfamily of Barnes (1958) and even less to that of Petrunkevitch (1928, 1939). Of the genera examined, I would retain only Marpissa and Maevia. Males of the genus Marpissa tend to exhibit courtships in which the opisthosoma is raised nearly perpendicular to the substrate. One species, M. sulcosa Barnes, differs from the others in twisting the opisthosoma to the right or left and exhibiting a zigzag motion with the prosoma elevated. It is the latter motion and elevation of the prosoma which seems to link M. sulcosa to Maevia. The genera are also similar in several morphological characteristics, such as in having four pairs of spines on the ventral first tibiae. These genera are separated by Proszyriski (1976), who restricts the subfamily to the genus Marpissa alone. Subfamily Aelurillinae: Menemerus bivittatus (Dufour) (R). Early Movements: direct or at slight angle. Later Movements: direct. Palpi: stationary. First Legs: partly raised, occasionally lowered during stationary periods. Prosoma: low. Opisthosoma: straight. Number of Observations: 12. Number of Males: 3. Number of Mountings: 7. Sexual Dimorphism: sexes differ primarily in pattern of dorsal surface, males with dark central stripe on opisthosoma, females with wide light band. Metacyrba californica (Peckham and Peckham) (R). Early Movements: direct, jerky. Later Movements: direct, jerky. Palpi: stationary. First Legs: raised, vibrated slightly. Prosoma: low. Opisthosoma: straight. Number of Observations: 4. Number of Males: 3. Number of Mountings: 3. Sexual Dimorphism: sexes very similar. M. undata (DeGeer) (R). Similar to M. californica except: Palpi: extended laterally, during early courtship stationary. First Legs: extended laterally during early courtship, raised to position like that of M. californica in late courtship, raised and lowered slightly during forward motion. Prosoma: raised. Opisthosoma: bobbed during mounting. Number of Observations: 2. Number of Males: 1. Number of Mountings: 1. Sexual Dimorphism: sexes very similar. Phlegra fasciata (Hahn) (I?). Early Movements: zigzag. Later Movements: direct. Palpi: not observed. First Legs: not recorded. Prosoma: not recorded, probably low. Opisthosoma: straight. Number of Observations: 1. Number of Males: 1. Number of Mountings: 0. Sexual Dimorphism: sexes similar, male with cephalic area brighter red than that of female; clypeus turquoise blue. The genera Menemerus, Metacyrba and Phlegra seem to be related, based on the similarities of their very simple courtships. Hill (1979) has found the opisthosomal scales of the first two genera to be nearly identical and unlike those of Marpissa. He has also divided the North American Metacyrba into two genera, Metacybra [M. taeniola (Hentz) and allied species] and Platycryptus (M. undata and allied species). I am inclined to agree RICHMAN-EPIGAMIC DISPLAY IN JUMPING SPIDERS 51 that this division is probably needed (based on morphology), although I have no data on the courtship of M. taeniola. I doubt whether these genera should be placed in the Aelurillinae, but they do not seem to belong to the Marpissinae, as thought by Petrunkevitch (1928) and Barnes (1958) or the “Habrocestinae” of Hill (1979). Phelgra has been placed in the Pelleninae (Petrunkevitch, 1928), but does not seem to belong there based either on courtship or on morphology (Proszynski 1976). Ethograms of the typical courtships of both Menemerus bivittatus and Metacyrba undata (Figures 1-2) indicate the relative simplicity of the displays in this group of genera. During the current study, males of M. bivittatus were observed to spend more time on areas of filter paper which had been in contact with virgin females than on clean filter paper (79% of 2-minute times periods on female-contacted paper, n = 10, SD = 1.8%, males = 2). This is taken as evidence for the existance of a possible contact pheromone. Other jumping spiders tested [Hentzia palmarum (Hentz) and Pellenes brunneus Peckham and Peckham] exhibited no preference. Menemerus bivittatus males were also observed engaged in both ritual agonistic display and fighting. Two males were sighted on the outside of a building in Gainseville, Florida. One male (the larger of the two) approached the other male, which was inside a silk retreat under a wooden beam. The second male eventually came out of its retreat after the first male started a side-stepping display. Both displayed with the first legs widely spread and nearly parallel to those of the other male. They fought, using their chelicerae at close range. At intervals they broke away from one another and the larger male finally drove the smaller male away, briefly occupying the abandoned silk retreat. It should be noted that Bhattacharya (1936) reported agonistic display for this species in India. Crane (1949) stated that M. bivittatus had no agonistic display. Subfamily Synemosyninae: Sarinda cutleri (Richman) (R). Early Movements: direct. Later Movements: with some indication of zigzag. Palpi: extended laterally, stationary. First Legs: raised, vibrated. Prosoma: raised. Opisthosoma: not observed. Number of Observations: 2. Number of Males: 2. Number of Mountings: 0. Sexual Dimorphism: Male with scale-covered elongated chelicerae. S. hentzi (Banks) (R). Early Movements: zigzag. Later Movements: direct? Palpi: extended laterally, stationary. First Legs: raised, wave in unison. Prosoma: raised. Opisthosoma: bobbed up and down. Number of Observations: 5. Number of Males: 2. Number of Mountings: 0. Sexual Dimorphism: male with scale-covered elongated chelicerae. Synemosyna formica Hentz (R). Early Movements: not observed. Later Movements: direct. Palpi: not observed. First Legs: raised. Prosoma, raised. Opisthosoma: bobbed up and down. Number of Observations: 1. Number of Males: 1. Number of Mountings: 0. Sexual Dimorphism: sexes very similar. The ant-mimicking salticids were all placed into the Synemosyninae by Proszynski (1976), but several objections can be raised about this treatment. The members of Sarinda and Synemosyna raise their first legs in courtship, whereas the members of the genera Synageles, Semorina and Peckhamia (see Crane 1949, Table II) have a quite different courtship approach, the first legs being lowered and the opisthosoma being raised vertically or moved from side to side. The genital morphology seems also to 52 THE JOURNAL OF ARACHNOLOGY separate these groups and to relate the former two genera with Myrmarachne. It thus is quite doubtful that ant mimicry arose only once as Proszynski would seem to indicate. Subfamily Sitticinae: Neon nelli Peckham and Peckham (I?). Early Movements: zigzag. Later Movements: zigzag. Palpi: raised and lowered in unison. First Legs: raised, moved in unison. Prosoma: only slightly raised. Opisthosoma: straight. Number of Observations: 2. Number of Males: 2. Number of Mountings: 0. Sexual Dimorphism: sexes very similar. Sitticus cursor Barrows (I?). Early Movements: direct. Later Movements: zigzag. Palpi: not observed. First Legs: first raised then lowered. Prosoma: only slight raised. Opisthosoma: straight. Number of Observations: 5. Number of Males: 2. Number of Mountings: 0. Sexual Dimorphism: sexes very similar, male slightly darker than female. There may be some difficulty with the placement of Sitticus and Neon in the Sitticinae (Proszynski 1976), but not enough is known about their relationship at present to settle the matter. Subfamily Thiodininae : Thiodina puerpera (Hentz) (I?). Early Movements: not well observed. Later Movements: not well observed. Palpi: lowered, stationary. First Legs: raised. Prosoma: raised. Opisthosoma: lowered. Number of Observations: 1. Number of Males: 1. Number of Mountings: 1 (mated with female T. sylvana). Sexual Dimorphism: sexes differ primarily in dorsal color pattern, male generally darker than female and with dark or banded legs. T. sylvana (Hentz) (I?). Early Movements: lateral or direct, very slow, jerky. Later Movements: not observed. Palpi: lowered, mostly stationary. First Legs: raised, then raised and lowered alternately or in unison, no apparent pattern. Prosoma: raised. Opisthosoma: lowered. Number of Observations: 2. Number of Males: 2. Number of Mountings: 0. Sexual Dimorphism: dimorphism as in T. puerpera , from which this species differs in details of dorsal pattern. While I am inclined to agree with Proszynski (1976) that Thiodina is closely related to the Dendryphantinae, I have kept its separation from the latter subfamily based on the unusual sensory hairs on the first pair of legs and differences in body scales (Hill 1979). However, based on the structure of the palpi and epigyna and on the courtship behavior it should be closely allied with the dendryphantines. Subfamily Dendryphantinae: Eris marginata (Walckenaer) (I). Early Movements: zigzag. Later Movements: zigzag? Palpi: lowered, stationary. First Legs: raised, jerked. Prosoma: low to slightly raised. Opisthosoma: lowered. Number of Observations: 2. Number of Males: 2. Number of Mountings: 0. Sexual Dimorphism: male with somewhat elongated and enlarged chelicerae, usually darker with more contrasting markings than female, dorsal markings different. Hentzia grenada (Peckham and Peckham) (I). Early Movements: often zigzag, sometimes straight. Later Movements: as in early. Palpi: extended laterally. First Legs: extended laterally, raised, lowered alternately and in unison several times. Prosoma: raised. Opisthosoma: twisted to right or left, straightened at intervals. Number of Observations: 25. Number of Males: 2. Number of Mountings: 6 (matings only with virgin female //. palmarum). Sexual Dimorphism: male with elongated chelicerae, body more elongate than female, clypeus of male white and first legs dark, dorsal markings different. RICHMAN-EPIGAMIC DISPLAY IN JUMPING SPIDERS 53 H. mitrata (Hentz) (I). Courtship as in H. grenada except front legs raised and lowered more often, usually with each forward movement. Number of Observations: 9. Number of Males: 2. Number of Mountings: 0. Sexual Dimorphism: male with fringe of white hair on first legs, first legs more elongate, dorsal markings different. H. palmarum (Hentz) (I). As in H . grenada. Almost impossible to distinguish the two species on the basis of courtship. Can hybridize with H. grenada. Number of Observations: 31. Number of Males: 11. Number of Mountings: 10. Sexual Dimorphism: dimorphism essentially as in H. grenada , from which this species differs in details of dorsal pattern. Metaphidippus galathea (Walckenaer) (I). Early Movements: zigzag. Later Movements: zigzag. Palpi: moved with no pattern. First Legs: lowered, bent upward slightly. Prosoma: low. Opisthosoma: straight. Number of Observations: 3. Number of Males: 2. Number of Mountings: 1. Sexual Dimorphism: sexes differ in color pattern, males with more distinct markings and with white scales and hairs on darker background color. M. manni (Peckham and Peckham) (I). Early Movements: zigzag. Later Movements: zigzag. Palpi: extended laterally, raised and lowered in unison while moving. First Legs: lowered extended laterally at wide angle at first, narrowing as male approaches female. Jerked at intervals. Prosoma: low or only slightly raised. Opisthosoma: straight. Number of Observations: 3. Number of Males: 3. Number of Mountings: 1. Sexual Dimorphism: sexes differ in color pattern, male with more distinct markings and with white stripes on clypeus and chelicerae. M. sexmaculatus (Banks) (I). Early Movements: direct. Later Movements: direct. Palpi: moved with no pattern. First Legs: raised, moved with no pattern. Prosoma: Low. Opisthosoma: straight. Number of Observations: 2. Number of Males: 1. Number of Mountings: 0. Sexual Dimorphism: sexes very similar. M. vitis (Cockerell) (I). Early Movements: zigzag, with some arc-like motion. Later Movements: as in early. Palpi: stationary. First Legs: raised, crossed, jerked at intervals. Prosoma: low. Opisthosoma: straight. Number of Observations: 3. Number of Males: 2. Number of Mountings: 0. Sexual Dimorphism: sexes very similar, males darker in color. Phidippus pulcherrimus Keyserling (I). Early Movements: zigzag. Later Movements: direct. Palpi: stationary. First Legs: raised, jerked. Prosoma: raised. Opisthosoma: straight. Number of Observations: 1. Number of Males: 1. Number of Mountings: 0. Sexual Dimorphism: male with black and white fringes on first legs; opisthosomal markings more distinct; palpi with contrasting white and black areas. Sassacus papenhoei Peckham and Peckham (I). Early Movements: zigzag or spiral. Later Movements: as in early. Palpi: not observed. First Legs: raised, crossed, raised and lowered later during lateral movements. Prosoma: low. Opisthosoma: straight. Number of Observations: 1. Number of Males: 1. Number of Mountings: 1. Sexual Dimorphism: sexes similar. Tutelina elegans (Hentz) (I?). Early Movements: zigzag. Later Movements: zigzag. Palpi: extended laterally, stationary. First Legs: raised, waved alternately or in unison (erratic). Prosoma: raised. Opisthosoma: straight. Number of Observations: 4. Number of Males: 1. Number of Mountings: 0. Sexual Dimorphism: sexes differ in color, males with grey scales on body, black fringe around eyes and with white tarsi on first legs, female iridescent green on dorsal surface. The genera in the Dendryphantinae are, in some ways, difficult to characterize. While the subfamily is distinct morphologically, it is behaviorally polythetic— no one courtship characteristic being common to all, or even most, genera. Yet, most of the genera are 54 THE JOURNAL OF ARACHNOLOGY related to one or more genera within the subfamily by one or more behavioral traits. For example, opisthosomal twisting is common in some genera ( Hentzia , Zygoballus, and some Phidippus) and one of these, Phidippus , is linked to Eris, Tutelina, and even Thiodina by its characteristic side-stepping hesitating movement, with first legs raised and prosoma elevated. The similarity of the courtships of Sassacus papenhoei and Metaphidippus vitis may indicate that these two species are congeneric, both belonging to Sassacus as proposed by Hill (1979). Males of both species raise and cross the first legs, lower the prosoma, jerk the first legs up and down at the start of their sideways runs and both have zigzag or spiral paths to the female. Zygoballus , although not studied in the current work, has both a similar morphology and similar epigamic displays to the dendryphantines (Peckham and Peckham 1889). I thus agree with Hill (1979) that the Zygoballinae of Petrunkevitch (1928) should become part of the Dendryphantinae. In morphology Zygoballus resembles Eris while in courtship and agonistic display it resembles Hentzia. The genus Hentzia is an example of a group of species which exhibits ecological separation, possible competitive exclusion, and apparent recent contact between closely related species. Hentzia palmarum (Hentz) seems to have been a long time resident in Florida, whereas H. grenada Peckham and Peckham may have relatively recently invaded Florida from the South America-West Indies area. The latter is primarily found on palms, especially the saw palmetto in the north. It has been collected as far north as the Okefenokee Swamp in Georgia, but is more abundant in the slash pine-palmetto areas of south Florida. H. palmarum has been collected in abundance on coastal mangroves, willow trees along lakes, Lyonia and scrub oaks in sand pine scrub and turkey oak-longleaf pine associations, and on various other shrubs. It is very rarely collected on palms, although it may be collected next to H. grenada in mixed palmetto and shrub understories. The two species have nearly identical courtship (Figures 3-5) and agonistic displays and will interbreed in the laboratory; virgin females of H. palmarum readily accepted males of H. grenada. In interspecific agonistic display-fights between equal sized males, the male of H. grenada always won. The resultant progeny of crosses between H. palmarum and H. grenada never matured (n = 2), but the sample is not large enough for any conclusions. Morphologically the two species are distinct, although they are closely related enough to be quite similar. Males of H. grenada appear to lack the variation in cheliceral length which has been observed in males of H. palmarum. All of the males of H. grenada collected in Florida and Georgia (approximately 20) had very long chelicerae, longer than for a male of H. palmarum of similar size. The courtship of both species is relatively complex, although in a few cases a male approached a female and mated with her after very little visual display. At the start of most of the displays the male was facing the female at a distance of 3-4 cm. Usually any movement of the female initiated a display by the male. He would then spread his first legs widely apart (approximatly 100°) and advance toward the female, usually with the opisthosoma twisted to the right or left and raised approximately 30°. Most males proceeded in a zigzag path toward the female, with pauses about twice a second, during which the opisthosoma was straightened and then twisted again. A few males (Figure 4) switched the opisthosoma from right to the left or left to right during the display, but most kept one orientation or the other when the opisthosoma was twisted. The magnitude of the zigzags varied from a nearly straight path to sideways motions of as much as 1 cm. When the male was within 1 cm of the female he usually straightened his abdomen and extended his first legs forward, toward the female. During the final stage of RICHMAN-EPIGAMIC DISPLAY IN JUMPING SPIDERS 55 courtship the male alternately raised and lowered his first legs two or three times, in addition to raising them slightly at the start of each forward movement. If the female was receptive (which she usually was, if virgin— of 12 virgin H. palmarum females, 10 accepted males) the male moved forward, alternately touching one first leg to the female and then the other in a see-saw manner approximately five times. He then mounted and twisted the female’s opisthosoma around so that he was able to insert one of his palps into one side of the epigynum, later repeating with the other palpus on the other side. Males often shifted from one side to the other several times, until the female started moving and apparently disengaged herself. Males removed from a female after mounting usually resumed copulation without repeating courtship. Both H. palmarum and H. grenada were observed to perform identical agonistic displays which often ended in fighting. The displays, which were observed eight times for H. palmarum and nine times for H. grenada (not counting interspecies displays) began with one or both males extending the chelicerae, raising the first legs slightly, and raising and twisting the opisthosoma. Both males moved forward and began a rocking motion in synchrony; the first legs were extended laterally and eventually were parallel to those of the other male. The fangs were extended and the males fought (this usually occurred even if the males were not evenly matched). Finally one broke away and ran. The displays were very short, usually lasting less than 15 seconds. Subfamily Euophryinae : Corythalia canosa (Walckenaer) (H). Early Movements: usually direct. Later Move- ments: zigzag. Palpi: rotated, one clockwise, other counter-clockwise in phase. First Legs: lowered, raised slightly during motion. Prosoma: raised. Opisthosoma: straight. Number of Observations: 11. Number of Males: 5. Number of Mountings: 1. Sexual Dimorphism: sexes similar, male with white scales on clypeus and chelicerae, palpi more contrastingly marked than those of female. Habrocestum bufiodes Chamberlin and Ivie (H). Early Movements: body moved back and forth by shifting position over legs, some zigzag motion. Later Movements: as in early. Palpi: raised and lowered alternately or in unison. First Legs: lowered. Prosoma: raised. Opisthosoma: straight. Number of Observations: 4. Number of Males: 2. Number of Mountings: 0. Sexual Dimorphism: sexes differ in color pattern, males with more contrastingly marked palpi; markings more distinct than those of female. H. pulex (Hentz) (H). Early Movements: zigzag or arc. body tilted toward direction of movements. Later Movements: zigzag or arc. Palpi: lowered, stationary, occasionally raised and lowered slightly in unison or while body tilted, one lowered and one raised. First Legs: lowered during early courtship; raised during later courtship. Prosoma: lowered early, raised later. Opisthosoma: straight. Note: female actively participates in display, usually causing male to move in arc or circle around her. Number of Observations: 2. Number of Males: 2. Number of Mountings: 0. Sexual Dimorphism: sexes very similar, males with somewhat more distinct pattern. H. n. sp. A (H). Early Movements: stationary periods, followed by rapid decreasing arc-zigzag run. Later Movements: as in early? Palpi: raised and lowered alternately or in unison. First Legs: lowered. Prosoma: slightly raised. Opisthosoma: straight. Number of Observations: 10. Number of Males: 2. Number of Mountings: 0. Sexual Dimorphism: Sexes differ in color pattern, males with bright red cephalic area and with more distinct dorsal pattern. 56 THE JOURNAL OF ARACHNOLOGY Neonella vinnula Gertsch (R). Early Movements: whole body vibrated. Later Movements: jumped at female (may be atypical). Palpi: stationary during vibration of body. First Legs: lowered. Prosoma: raised at least slightly. Opisthosoma: straight. Number of Observations: 3. Number of Males: 1. Number of Mountings: 0. Sexual Dimorphism: sexes very similar. The genera examined in the current study ( Corythalia , Habrocestum and Neonella) seem to be related by their courtship. This relationship is especially evident between some species of Corythalia and Habrocestum. Proszyriski (1976) has placed all three genera into the Euophryinae and has presented convincing evidence of the relationship of these with each other and with Euophrys, based on genital morphology. Despite the differences in the opisthosomal scales found by Hill (1979) between Habrocestum and Corythalia , I find other morphological as well as behavioral evidence to favor Proszynski’s classification. I thus do not consider that Hill’s “Habrocestinae” is a valid subfamily. Two of the species of Habrocestum are closely related. These are H. bufoides and an undescribed species from the Florida scrub and pine forests. These species are very similar, superficially. The genitalia are distinct and their courtships are even more so (Figures 6-7). Males of H. bufoides perform a relatively slow rocking display, whereas males of H. n. sp. move in rapid arcing runs, the palpi being raised alternately at the end of each run. The motion is so fast that slow motion at one-half normal speed was needed so that the movies could be analyzed. The species are also somewhat ecologically isolated, although they were occasionally found together in ecotones. H. bufoides was abundant in mesic hammocks and pine flatwoods in north Florida, whereas H. n. sp. was collected in turkey oak-longleaf pine and sand pine scrub associations. The full description of the new species will be presented in a planned revision of Habrocestum. None of the species of Habrocestum was observed to exhibit any kind of agonistic display. Courtship in Corythalia canosa (Walckenaer) (Figure 8) is similar to that of C. fulgipedia (Crane 1948:28) in that the male’s first legs are kept down on the substrate and the palpi are rotated in a clockwise-counter clockwise fashion (the left palpus was observed in one example to be moving clockwise, while the right palpus moved counter-clockwise). The male courtship usually began approximately 3 cm from the female. The first legs of the male were extended laterally and the palpi were in constant motion at the rate of 3.33 turns per second. The path of the male showed considerable variation, being zigzag to nearly direct. The male moved forward in jerky motions, with his body somewhat elevated off the substrate and the opisthosoma straight. Females usually responded to the approach of the male by raising the first legs and elevating the prosoma (also seen in Habrocestum pulex and some Pellenes). All females observed either exhibited this behavior or ran from the male. At this point, with the female approximately 1 cm away, the male usually zigzaged back and forth, with his palpi in motion, his first three pairs of legs extended laterally and nearly parallel to each other, and his body elevated. The male then attempted to mount, raising his first legs upward and elevating his prosoma higher. One male finally managed to mount over the female’s elevated prosoma, and successfully mated, inserting his emboli alternately into the female’s epigynum. C. canosa differed from Crane’s (1948) South American species in that the agonistic displays of males often led to fighting if the males were of similar size. The agonistic display began with a male elevating its prosoma. The opisthosoma was either straight or bent downward and the first three pairs of legs were nearly parallel to each other and were touching the substrate. Either both males approached one another or one initiated RICHMAN-EPIGAMIC DISPLAY IN JUMPING SPIDERS 57 the approach. As in courtship the palpi were in motion most of the time, the movements being mostly up and down, rather than in circles. The approach of the male was jerky, with very short pauses after forward runs. One male was observed to raise his third pair of legs off the substrate while he was under attack. Before contact was finally make in the encounters observed, one or both of the males raised his first legs. Usually one male raised his first legs and the other followed, but this was not dependent on which individual initiated the display. If one of the males had not been driven off at this point, contact was made and the first legs of both males were lowered and held parallel to those of the opponent (similar to the fighting poses of Menemerus and Hentzia). They fought, pushing each other back and forth and seeming to bite with their chelicerae. One of the males eventually broke away and ran. No fighting took place in one confrontation between males observed in nature. The smaller of the two males retreated after displaying for only a few seconds. Males of C canosa often differed markedly in size. It is doubtful that fights often take place between smaller and larger males as the smaller usually retreated in observed laboratory encounters as well. A total of 30 interactions between five marked males were observed in the laboratory to determine if the same males succeeded in driving away their opponents. In all cases the larger males defeated smaller males, usually without fighting. Males of nearly the same size as their rival fought until one eventually defeated the other. Repeated fights between such males always resulted in the same male “winning.” These encounters were observed over three days. Neonella, while having some morphological and behavioral similiarities to the euophryines, may actually belong with Neon in the Sitticinae. I have too little data at present to be sure of its placement. Subfamily Pelleninae: Pellenes agilis (Banks) (H). Early Movements: zigzag. Later Movements: direct. Palpi: stationary. First Legs: bowed downward, then raised. Prosoma: slightly raised. Opisthosoma: bobbed up and down. Number of Observations: 3. Number of Males: 2. Number of Mountings: 0. Sexual Dimorphism: male with long fringes on first legs, dorsal color pattern differs from that of female. P. arizonensis Banks (H). Early Movements: zigzag. Later Movements: direct. Palpi: lowered, some movement in unison. First Legs: bowed downward, then extended laterally and later forward, touching female. Prosoma: slightly raised to low. Opisthosoma: raised slightly, bobbed. Number of Observations: 11. Number of Males: 7. Number of Mountings: 1. Sexual Dimorphism: dimorphism essentially as in P. agilis, from which this species differs in details of dorsal pattern. P. brunneus Peckham and Peckham (H). Early Movements: zigzag. Later Movements: stationary for long periods of time, may arc around female. Palpi: move back and forth, raised and lowered in unison early, later extended forward. First Legs: raised, tarsi jerked at intervals. Third Legs: raised and lowered. Prosoma: low. Opisthosoma: straight, bob- bed slightly. Number of Observations: 32. Number of Males: 3. Number of Mountings: 0. Sexual Dimorphism: male with two prominent spatulate spines on prolateral surface of each first leg, third legs modified with projection on each patella and knobby distal end to femora; latter with dark spots and stripes; dorsal color pattern differs from that of female, male darker. P. calcaratus Banks (H). Early Movements: direct. Later Movements: direct. Palpi: touch substrate alternately during part of sequence with first legs bent downward, other- wise stationary. First Legs: raised at start of sequence, bends legs downward at tibia- metatarsus joint and vibrates them slowly upward, legs suddenly straightened, then 58 THE JOURNAL OF ARACHNOLOGY vibrated slowly downward and quickly raised ca. 6 times— sequence repeated. Third Legs: raised, lowered at intervals. Prosoma: low, lowered further during lowering of first legs. Opisthosoma: straight. Number of Observations: 3. Number of Males: 2. Number of Mountings: 0. Sexual Dimorphism: males with greenish first legs with white fringes and spatulate spines; third legs modified with spike on each patella; dorsal color pattern differs from that of female. P. carolinemis Peckham and Peckham (H). Early Movements: direct. Later Move- ments: direct. Palpi: lowered, stationary. First Legs: raised, waved in unison at intervals. Prosoma: raised slightly. Opisthosoma: bobbed up and down. Number of Observations: 8. Number of Males: 4. Number of Mountings: 0. Sexual Dimorphism: sexes differ in dorsal color pattern, males with more distinct pattern, usually darker. P. clypeatus (Banks) (H). Early Movements: somewhat zigzag, disorganized. Later Movements: as in early. Palpi: lowered. First Legs: raised, then raised and lowered in unison. Prosoma: lowered. Opisthosoma: straight. Number of Observations: 1. Number of Males: 1. Number of Mountings: 0. Sexual Dimorphism: male with spatulate spines on first legs: stripes on clypeus; dorsal pattern differs from that of female, male darker. P. coecatus (Hentz) (H). As in P. brunneus. Almost identical. Number of observations: 7. Number of Males: 2. Number of Mountings: 0. Sexual Dimorphism: male with long greenish fringes and spatulate spines on first legs; third legs modified with projection on each patella; clypeus red; white spots on chelicerae; white areas lateral to clypeus. P. cf. coecatus (H). Courtship similar to P. brunneus , except that the palpi are lowered and mostly stationary. Number of Observations: 17. Number of Males: 11. Number of Mountings: 1. Sexual Dimorphism: similar to that of P. coecatus but male with different shaped projections on third patellae; no white spots on chelicerae; no white lateral to clypeus. P. hallani Richman (H). Early Movements: zigzag. Later Movements: zigzag, in both early and later stages makes short, explosive jumps laterally or forward the width of the body at each of lateral movements. Palpi: lowered, mostly stationary or raised and lowered slightly in unison. First Legs: raised, occasionally raised and lowered in unison. Second Legs: displayed by bending backwards, exposing iridescent femora. Prosoma: low. Opisthosoma: straight. Number of Observations: 10. Number of Males: 3. Number of Mountings 1. Sexual Dimorphism: male covered with iridescent scales; clypeus iridescent pearl; femora of first two pairs of legs flattened or concave anteriorally and iridescent; dorsal color pattern more distinct than in female. P. hirsutus (Peckham and Peckham) (H). Early Movements: zigzag. Later Movements: not observed. Palpi: not observed. First Legs: raised, lowered alternately. Prosoma: not recorded. Opsithosoma: raised. Number of Observations: 5. Number of Males: 3. Number of Mountings: 0. Sexual Dimorphism: male with long black fringes and iridescent blue metatarsi on laterally flattened first legs; prosoma often with reddish or golden cephalic area; body covered with iridescent metallic scales; much darker in color than female. P. tarsalis Banks (H). Early Movements: zigzag. Later Movements: zigzag. Palpi: move back and forth in unison, then alternately, then held forward. First Legs: raised and bowed, tarsi jerked. Prosoma: low. Opisthosoma: straight. Number of Observations: 9. Number of Males: 6. Number of Mountings: 5. Sexual Dimorphism: male with longer front legs than female; tarsi of first legs black; palpi black and white; dorsal pattern of male differs from that of female; male darker. P. trimaculatus (Bryant) (H). Early Movements: zigzag. Later Movements: zigzag. Palpi: extended, moved alternately or in unison. First Legs: raised. Prosoma: low. RICHMAN-EPIGAMIC DISPLAY IN JUMPING SPIDERS 59 Opisthosoma: straight. Number of Observations: 12. Number of Males: 4. Number of Mountings: 0. Sexual Dimorphism: sexes similar except that male has different cephalic markings (white stripes in eye region) than female (dark bands on clypeus); male color pattern more distinct. P. viridipes (Hentz) (H). Early Movements: direct. Later movements: direct. Palpi: touch substrate nearly in unison as first legs are lowered, otherwise stationary. First Legs: raised, then slowly lowered while being vibrated back and forth and bending downward and inward at the tibia-metatarsus joint, suddenly jerked upward, sequence repeated. Third Legs: raised, lowered at intervals, not correlated to other movements, may be alternate or in unison. Prosoma: low. Opisthosoma straight, bobbed at intervals. Number of Observations: 3. Number of Males: 2. Number of Mountings: 0. Sexual Dimorphism: male with greenish first legs with long white fringes and spatulate spines; third legs modified with triangular patellae; male dorsal color pattern differs from that of female; dorsal color pattern very similar to that of P. calcaratus. P. cf. viridipes (H). Early Movements: not well observed. Later Movements: not observed. Palpi: lowered, stationary. First Legs: raised. Prosoma: low. Opisthosoma: straight. Number of Observations: 1. Number of Males: 1. Number of Mountings: 0. Sexual Dimorphism: dimorphism similar to that of P. viridipes, but third legs of male not modified. P. n. sp. A (H). Early Movements: direct. Later Movements: direct, almost motionless for long periods. Palpi: not observed. First Legs: raised. Prosoma: low. Opisthosoma: straight. Number of Observations: 2. Number of Males: 2. Number of Mountings: 0. Sexual Dimorphism: male with long golden-yellow fringes on first legs; clypeus bright red; dorsal color pattern more distinct than that of female. Pellenes, while morphologically distinct, is such a diverse genus with respect to court- ship that it is difficult to characterize. It seems best to retain the Pelleninae for this genus while noting the difficulties involved. Courtships among members of a particular species group were similar (see Griswold 1976). Thus members of, e.g., the agilis group (. P . agilis, P. arizonensis and possibly P. hirsutus) have similar courtships to fellow species-group members, but differed markedly from other species. Although they are distinct morpho- logically, the members of the viridipes group (P. calcaratus, P. viridipes and P. cf. viridipes) had courtships somewhat similar to those of the members of the coecatus group (P. brunneus, P. coecatus and P. cf. coecatus). The one member of the americanus group for which courtship was observed, P. tarsalis, had a completely distinct courtship from the other species studied, which involved the bowing of the first legs and jerking of the tarsi. P. hallani , which is a member of the umatillus group, also differed from all other species in that the male jumped its own width at the end of each zigzag, displaying its iridescent first and second femora. Pellenes is another genus where courtship may be useful to distinguish between closely related species. This is especially true for separating P. calcaratus fromP. viridipes and P coecatus from P. cf. coecatus. The latter appears to be a very abundant, but overlooked, undescribed species from the western United States. It has been confused withP. signatus (Banks) and more recently withP. coecatus. Ethograms are shown for the courtship of individual males of P. calcaratus and P viridipes (Figures 9-10), and P. brunneus, P. coecatus andP cf. coecatus (Figures 11-13). In the courthsip of P. calcaratus (Figure 9) the first legs were bent at the tibia- metatarsus joint and vibrated slowly upward at the start, followed by a sudden straighten- ing of the legs and then a series of approximately eight slow lowerings of the rapidly 60 THE JOURNAL OF ARACHNOLOGY vibrating first legs. At the end of each lowering the first legs were suddenly raised again. The palpi were moved only during the early raising of the vibrating and bent first legs. The palpi were raised and lowered alternately at a rate of 1.25/second for 4 seconds. In the courtship of P. viridipes (Figures 10) the male’s first legs were bent at the tibia- metatarsus joint, both downward and inward. The legs were never raised while so bent, nor were they lowered at a steady rate. Instead they were at times almost stationary in height while the tarsi were rotated in a small circle. The first legs were finally lowered the rest of the way and suddenly straightened and raised. This sequence was repeated. During the initial movements of the first legs the palpi were moved up and down rapidly nearly in unison at a rate of approximately 2.7/second for 1.5 seconds. The third legs of both species were moved up and down and back and forth at intervals, apparently with little pattern. The movements of the first legs and the palpi serve to separate these remarkably similar species (the dorsal patterns of the males are nearly identical). A female of P. calcar atm was used to obtain courtship from P. viridipes males because no P. viridipes females were available. It is unlikely that the male varies the basic movements of the courtship, even in such an artificial situation. The courthsip observed for P. viridipes matches a published account by Peckham and Peckham (1890). No distinctive agonistic display was observed. The courtship of Pellenes brunneus (Figure 11) usually began when the male was approximately 3 cm from the female. The male initially raised his first legs and spread laterally his palpi as widely as possible. He usually moved in a zigzag during early court- ship, switching later to an arc around the female. During early courtship the palpi were raised and lowered at a rate of approximately 3/second for for 5-10 seconds, while still being kept widely separated. The male moved forward toward the female while raising and lowering his first legs and pausing several times until he was approximately 0.5 cm from the female. The male then assumed a characteristic pose of members of the Pellenes coecatus species groups, with the first legs held high, the palpi lowered, extended forward and held parallel and stationary, and the body low to the substrate. The third legs were pressed to the sides of the opisthosoma. During this pose the male often remained stationary for 15-30 minutes, especially if the female was not turned toward him. If the female turned toward the male, he raised and lowered his first tarsi very rapidly several times and this activity was often preceded or followed by the raising of the third legs, which were also moved backward and forward. The third legs were usually raised one at a time and the male ended the sequence by touching both third patellae together over his dorsal surface and then lowering them to their original positions. Only one mounting was observed and this was with a chilled female. In this case the male continued his display for several minutes, creeping slowly forward until he was able to touch the female. The male then leaned forward and touched the female with one of his first legs several times. This activity was repeated alternately with the right and left legs. The third legs were moved again several times, the last time just before mounting. The male then mounted and attempted to mate as in other salticids. Virgin females of P. brunneus were difficult to obtain and all females seemed to be quite resistant to male advances. This seemed to be true with most other species of Pellenes and it is possible that they required conditions which were not present in the laboratory. The display of P. coecatus (Figure 12) was very similar to that of P. brunneus , within variation shown by that species, but differed from P. cf. coecatus (Figure 1 3) in that the palpi of the eastern form (P. coecatus) were extended laterally and moved up and down during early courtship, whereas the western form (P. cf. coecatus) males kept their palpi RICHMAN-EPIGAMIC DISPLAY IN JUMPING SPIDERS 61 lowered and relatively stationary. The eastern form males differed from those of the western form also in coloration, the chelicerae having white patches, the areas lateral to the clypeus being white and the dorsal pattern also being somewhat different. Only one species of Pellenes was observed to have an agonistic display and this was P. carolinensis Peckham and Peckham [= P. tachypodus (Chamberlin and Ivie)] . The display was quite similar to courtship, except that the males pushed each other back and forth. This display was observed on three occasions in the laboratory. Subfamily Plexippinae: Plexippus paykulli (Audouin) (H). Early Movements: zigzag. Later Movements: direct. Palpi: not observed. First Legs: raised. Prosoma: raised. Opisthosoma: straight. Number of Observations: 3. Number of Males: 3. Number of Mountings: 0. Sexual Dimorphism: male with longer first legs than female; clypeus striped black and white; dorsal color pattern more distinct than that of the female. Plexippus has a rather simple courtship. It is certainly not related to the rapidly moving Corythalia as suggested by Crane (1949). The subfamily Plexippinae is retained for this genus mainly because it does not seem to fit elsewhere. The subfamily may be much more important in the Old World. CONCLUSION Courtship behavior in salticids can, with reservations, be utilized for systematic work. This can be at the species, genus or subfamily level, although the latter should be approached with the most caution. The separation of closely related species in Habroces- tum and Pellenes are examples of the first. The courtship similarities between species, such as in the genera Hentzia and Corythalia can help define genera, as an example of the second. Finally the similarities and differences between genera can help to define sub- families, such as the Euophryinae and the Dendryphantinae. Courtship behavior should be utilized primarily as supporting evidence for classifications or taxon determinations based on morphology, especially on reproductive structures. Many difficulties lie in determining just which characters, both morphological and behavioral, are synapomorphic (shared derived) and which are symplesiomorphic (shared ancestral). The similar courtships found in the Euophryinae and in members of genera such as Hentzia are possible examples of the first and certain widely scattered courtship characters such as the raising of the prosoma may be examples of the second. I believe that with care distinctions can be made, but a thorough knowledge of the species involved must be obtained before these can be reasonably certain. The adaptive radiation of the salticids has left the systematist with a complex pattern of parallel and convergent evolution. Unfortunately, there is little fossil evidence to help interpret this pattern. Thus it can only be worked out through indirect evidence, including the use of courtship behavior and morphology. The older subfamilial classifica- tions have become, for the most part, unworkable. Future classification systems will not be based on carapace shapes, eye positions and retromarginal teeth, which are often adaptations to a specific environment, but on reproductive morphology, behavior and ecology. 62 THE JOURNAL OF ARACHNOLOGY ACKNOWLEDGMENTS I would like to thank Drs. Jonathan Reiskind, John Anderson, Clifford Johnson, D. H. Habeck and James Lloyd of the University of Florida; W. B. Miller, P. E. Pickens and A. R. Mead of the University of Arizona; W. J. Gertsch of Portal, Arizona; and Mr. Vincent Roth of the American Museum Southwestern Research Station for their help with various parts of this study. Thanks also are due Dr. G. B. Edwards, Dr. David E. Hill, Dr. Bruce Cutler, Mr. Wayne Maddison and Mr. Will Kopachik for specimens and occasional advice. Special thanks go to Ms. Lynda Goin for her encouragement during the Florida part of the research. The project was in part funded through a Grant-in- Aid of Research from Sigma Xi. Specimens of all the species utilized in this study have been deposited in the Florida State Collection of Arthropods (FSCA), Gainesville, Florida. LITERATURE CITED Barnes, R. D. 1958. North American jumping spiders of the subfamily Marpissinae (Araneae, Salt- icidae). Amer. Mus. Novitates, 1867, 50 pp. Bhattacharya, G. C. 1936. Observations on some peculiar habits of the spider ( Marpissa melano- gnathus). J. Bombay Natur. Hist. Soc., 39:142-144. Bristowe, W. S. 1941. The comity of spiders. Vol. II. Ray Soc., London, 560 pp. Crane. J. 1948. Comparative biology of salticid spiders at Rancho Grande, Venezuela. Part 1. System- atics and life histories in Corythalia. Zoologica, 33: 1-38. Crane, J. 1949. Comparative biology of salticid spiders at Rancho Grande, Venezuela. Part 4. An analysis of display. Zoologica, 34:159-214. Griswold, C. 1976. Biosystematics of Habronattus in California. M.S. Thesis, Univ. of California, Berkeley, 187 pp. Hegdekar, B. M. and C. D. Dondale. 1969. A contact sex pheromone and some response parameters in lycosid spiders. Canadian J. Zool., 41:1-4. Hill, D. E. 1979. The scales of salticid spiders. Zool. J. Linnean Soc. London, 65:193-218. Jackson, R. R. 1976. The evolution of courtship and mating tactics in a jumping spider, Phidippus johnsoni (Araneae, Salticidae). Ph.D. Thesis, Univ. of California, Berkeley, 271 pp. Jackson, R. R. 1977. Courtship versatility in the jumping spider , Phidippus johnsoni (Araneae: Salt- icidae). Amin. Behav., 25:953-957. Jackson, R. R. 1978. An analysis of alternative mating tactics of the jumping spider Phidippus johnsoni (Araneae, Salticidae). J. Arachnol., 5:185-230. Land, M. F. 1969. Movements of the retinae of jumping spiders (Salticidae: Dendryphantinae) in response to visual stimuli. J. Exp. Biol. 51:471-493. Peckham, G. W. and E. G. Peckham. 1889. Observations on sexual selection in spiders of the family Attidae. Occ. Pap. Wisconsin Natur. Hist. Soc., 1:3-60. Peckham, G. W. and E. G. Peckham. 1890. Additional observations on sexual selection in spiders of the family Attidae, with some remarks on Mr. Wallace’s theory of sexual ornamentation. Occ. Pap. Wisconsin Natur. Hist. Soc., 1:117-151. Petrunkevitch, A. 1928. Systema Araneorum. Trans. Connecticut Acad. Sci., 29:1-270. Petrunkevitch. A. 1939. Catalogue of American spiders. Part One. Trans. Connecticut Acad. Sci., 33:133-338. Platnick, N. 1971. The evolution of courtship behavior in spiders. Bull. Brit. Arachnol. Soc., 3:40-47. Proszynski, J. 1971. Problems of classification of Salticidae (Aranei). Proc. 5th Internat. Arachnol. Congr., Brno., 213-217. Proszynski, J. 1976. Studium systematyczno-zoogeograficzne nad rodzina Salticidae (Aranei) Re- gionow Palearktycznego i Nearkycznego. Wyzsza Szkola Pedagogiczna W Siedlacach Rozprawy NR 6, 260 pp. RICHMAN-EPIGAMIC DISPLAY IN JUMPING SPIDERS KEY FOR SYMBOLS USED IN FIGURES 1-13 Direction A/V -/ h of path (DR) Path direct Path zigzag Path -an arc Movement forward Movement backward Stationary Break Chelicerae (CM) Chelicerae in normal resting position C Chelicerae extended, fangs not exposed Palpi (PA) -V- Palpi raised at angle — Right palpus raised Left palpus raised Palpi lowered at angle — *— Palpi extended laterally Palpi extended forward at angle Palpi extended directly forward Palpi moved up and down in unison M Palpi moved up and '' down alternately Palpi moved in circular path, one clockwise, other counterclockwise First legs (FL) \£- First legs raised at angle Right first leg raised CD Left first leg raised Second legs (SL) 7A First legs lowered at angle Second legs extended forward jLL First legs held Third legs (TL) perpendicular to substrate f/v\l I Third legs raised in unison First legs extended laterally 'f/W Third legs lowered in unison First legs extended forward at angle aa't Right third leg raised v Left first leg raised. Left third leg raised both extended forward *AA Left third leg lowered at angle Second and third pairs of legs (ST) First legs extended directly forward =*= Second and third legs nearly parallel First legs moved up and down in unison Prosoma (PR) X* First legs moved up rr Prosoma raised and down alternately n Prosoma moderately raised az First legs vibrated ** Prosoma lowered upward b Prosoma tilted to right AZ' First legs vibrated downward f Prosoma tilted to left First legs slightly Opisthosoma (OP) curved downward and HB Opisthosoma straight forward (normal posture in most -U First legs bowed inward species) v: First legs bent at tibia- metatarsus joint A Opisthosoma raised and twisted to right 1L First legs jerked suddenly upward (straightened) W Opisthosoma raised and twisted to left Tarsi and metatarsi of Opisthosoma lowered first legs jerked up and down % Opisthosoma bobbed (5 Tarsi of first legs rotated in circle while * Mounting legs are lowered Time (seconds) 63 Fig. 1.— Ethogram of courtship of Menemerus bivittatus male from Gainesville, Alachua County, Florida. 64 THE JOURNAL OF ARACHNOLOGY Fig. 2.— Ethogram of courtship of Metacyrba undata male from Charlton County, Georgia. Figs. 3-4.— Ethograms of courtship of two different Hentzia palmarum males from Way Key, Levy County, Florida. RICHMAN-EPIGAMIC DISPLAY IN JUMPING SPIDERS 65 Time (seconds) Time (seconds) I 6 £ 5 “94 g 3 X 2 — , * VVI-YV^h 7 Time (seconds) Fig. 5.— Ethogram of courtship of Hentzia grenada male from Archbold Biological Research Station, Highlands County, Florida. Female chilled. Fig. 6.— Ethogram of courtship of Habrocestum bufoides male from Lake Oklawaha, Putnam County, Florida. Fig. 7.— Ethogram of courtship of Habrocestum new species male from Gainesville, Alachua County, Florida. (cm) 66 THE JOURNAL OF ARACHNOLOGY Fig. 8.— Ethogram of courtship of Corythalia aanosa male from the River Styx, Alachua County, Florida. Fig. 9. -Ethogram of courtship of Pellenes caloaratus male from the Muskoka District, Ontario, Canada. Fig. 10. -Ethogram of courtship of Pellenes viridipes male from the Muskoka District, Ontario, Canada. RICHMAN-EPIGAMIC DISPLAY IN JUMPING SPIDERS 67 Time (seconds) Fig. 11.— Ethogram of courtship of Pellene6 brunneus male from Gulf Coast of Levy County, Florida. Female chilled. Fig. 12.— Ethogram of courtship of Pel lenes coecatus male from Jekyll Island, Glynn County, Georgia. Fig. 13.— Ethogram of courtship of Pellenes of. coecatus male from Yuma, Yuma County, Arizona. Hoffmaster, D. K. 1982. Predator avoidance behaviors of five species of Panamanian orb-weaving spiders (Araneae; Araneidae, Uloboridae). J. Arachnol., 10:69-73. PREDATOR AVOIDANCE BEHAVIORS OF FIVE SPECIES OF PANAMANIAN ORB-WEAVING SPIDERS (ARANEAE; ARANEIDAE, ULOBORIDAE) Debra K. Hoffmaster Department of Biological Sciences Texas Tech University Lubbock, Texas 79409 ABSTRACT The responses of five tropical orb-weaving spiders (family Araneidae and Uloboridae) to predatory attacks by salticids and hummingbirds were observed. Responses varied between species and predators. Salticid attacks dorsal to orb-weavers on the web evoked rapid movements which displaced the salticid from the orb-weaver’s dorsum. Ventral salticid attacks were countered by stilting. Attacks by hum- mingbirds evoked dropping or retreat from the web surface. Uloborus republicans (Simon) without eggmasses dropped more frequently than U. republicans with eggmasses when attacked by humming- birds. Cyclosa caroli (Hentz) did not respond to hummingbird attacks directed toward the debris-like stabilimentum; however, salticid spider contacts on the Cyclosa or stabilimentum evoked active de- fense. INTRODUCTION Tolbert (1975) described the predator avoidance behaviors of the orb-weaving spiders Argiope aurantia (Lucas) and Argiope trifasciata (Forskal) using an artificial model. The use of models as stimuli to elicit behavioral responses has obvious drawbacks when the sign stimuli or recognition patterns are not known. Nevertheless, Tolbert (1975) provided an important first look at predator avoidance patterns in araneids. This study examines the predator avoidance patterns of four species of tropical araneids and one uloborid. Two types of predators were examined; ambulatory and aerial predators. METHODS Four species of araneid and one uloborid were used as prey in staged predatory interactions involving salticid (Araneae) and hummingbird (Trochilidae) predators. The prey spiders were collected from second growth areas along Pipeline Road, near Gamboa, Panama. Spiders were returned to Barro Colorado Island, Panama, measured, and released into one of two adjoining 3 x 3 x 2 m insectary rooms. A fruit fly culture ( Drosophila sp.) was maintained in each insectary room to provide food for the spiders. The response of orb-weaving spiders to salticid predation was studied using the large salticid Phiale guttata (C. L. Koch), captured from the Pipeline Road collecting sites. Salticids were maintained in 3 x 10 mm plastic vials, corked with foam rubber. Salticids 70 THE JOURNAL OF ARACHNOLOGY were watered daily but fed infrequently to maintain their responsiveness to prey (Gardner 1964). Predatory encounters were staged by placing a slaticid on a 1.5 cm wooden platform which was held 3-5 cm from the intended prey. This method is consistent with records of salticid-araneid encounters (Robinson and Valerio 1977). The species of the intended prey was recorded, along with the point at which the salticid contacted the web or intended prey, the response of the intended prey, and the outcome of the encounter. The response of orb-weaving spiders to hummingbird predation was studied using the hermit hummingbird, Phaethornis superciliosus (Linne). Hummingbirds were mistnetted on Barro Colorado Island, Panama, placed in an insectary, and fed a 20% sugar solution without protein for two days prior to release into the insectaries containing spiders. All observations were made from within the insectaries. In all cases, my presence had little effect upon the activities of the predator or prey. A three-way test of independence (G test, Sokal and Rohlf 1969) was used to examine the independence of predator (hummingbird or salticid), prey species, and behavior. A Fisher exact test (Sokal and Rohlf 1969) was used to examine the differential response of uloborids with and without an eggmass. RESULTS AND DISCUSSION The three way interaction of prey species, predator species, and behavior is statistically significant (G test, P< 0.01). This indicates that predator avoidance behaviors vary, depending upon the species of araneid or uloborid attacked, and the predator. The apparent selection of different prey by hummingbirds and salticids may be an artifact of the experimental design, since salticids were not allowed to forage freely, nor were they offered the entire range of prey. Phiale guttata (C. L. Koch) (Family Salticidae).-The results of staged salticid-araneid interactions are presented in Table 1. Because of the artificial nature of the experiment, it is difficult to determine whether all salticid jumps were the result of predator attacks. Since the presentation of unoccupied platforms did not affect the behavior of the in- tended prey, I assume that the behavioral responses to salticid jumps reflect normal predator avoidance patterns, rather than abnormalities induced by the experimental method. Three attacks involving immature Nephila clavipes (Linne) resulted in kills. In each case the salticid jumped onto the Nephila\ abdomen and bit the Nephila immediately behind the cephalothorax. The salticid carried the prey to the platform via a dragline secured prior to the attack, and consumed the prey. The position of the salticid’s impact with the web or araneid has a strong effect (G test P < 0.5) on the responses of Argiope argentata (Frabricius) and Nephila clavipes (Table 1). Prey capture behaviors occurred in 72% of the instances when the salticid contacted the catching spiral of the web (n = 36, Table 1). In A. argentata, rapid lateral movements comprised 68% of the response to dorsal body contact (n = 29, Table 1). Pumping occurred in 16% of ventral contacts and 16% of catching spiral contacts. In N. clavipes, rapid lateral movements accounted for 100% of the response to dorsal body contact. Pumping occurred only once, in response to contact on the catching spiral (n = 11, Table 1). In both A. argentata and N. clavipes, stilting occurred only in response to ventral hub contacts. Responses of the cryptic species, Cyclosa caroli (Hentz) to contact by salticids are presented in Table 1. Pumping and lateral movements comprised 75% of the response to HOFFM ASTER- ARANEID PREDATOR AVOIDANCE BEHAVIORS 71 Table l.-The responses of orb-weaving spiders to predation by the jumping spider Phiale guttata. Unless otherwise indicated, all orb-weaving spiders were adult females. D = dorsal presentation; V = ventral presentation; CS = catching spiral. Behaviors Species No Response Prey Capture Pump Rear Stilt Move Drop Argiope argentata D 2 1 3 19 2 V 3 2 1 5 1 CS 2 15 4 3 Nephila clavipes D 11 V 7 3 1 CS 11 1 Cyclosa caroli 2 6 6 2 Uloborus republicanus 2 1 9 1 salticid contact on the araneid or the debris-like stabilimentum (n = 16). Attacks and drops were equally common, accounting for the remaining 25% of the response. The spider Uloborus republicanus (Simon) keeps the eggmass in the web, attached to the spider by a dragline issuing from the spinnerets. Rapid lateral movement was the most common response (69%) to salticid contacts. Uloborids with eggmasses carried the egg- mass, attached to the spinnerets, away from the predator when moving. Uloborids with eggmasses did not drop, an act which would require abandoning the eggmass. Uloborids without eggmasses dropped 20% of the time (n = 13). No response was a component of the defense of uloborids with eggmasses only. This trend was not significant (Fisher exact test, P = 0.35). The prevalence of stilting in response to ventral attacks involving Argiope argentata correlates with Tolbert’s (1975) report of predator avoidance in A. aurantia and A. trifasciata. The incidence of prey capture behaviors in response to salticid contact on the catching spiral probably relfects a predatory, rather than defensive, response (Robinson and Robinson 1973, Robinson and Olazarri 1971). Rapid lateral movements, switching, etc., dislodge the salticid before it can bite the prey. Dropping, seen infrequently in Tolbert’s (1975) study, was a component of salticid contacts on or near the orb-weaver itself. Dropping is common in the field, often occurring upon the observer’s approach. Dropping removes the spider from the web, a visual stimulus for aerial predators (Ed- munds 1974); however, dropping may bring the spider into contact with predators on the vegetation. Phaethomis superciliosus (Linne) (Family Trochilidae).-Hermit hummingbirds are known to forage on orb-weaving spiders (Young 1971). Foraging is characterized by low flights and examination of the web area. Attacks are usually dorsal, directed toward small spiders (body length < 7 mm), kleptoparasites, eggmasses and insects in the web. Forty percent of all attacks results in prey captures; however, in 60% of the attacks (n = 152), the hummingbird did not capture a spider in its first lunge. These instances were used to examine the spider’s antipredator behaviors. Uloborus republicanus and second instar Nephila clavipes (body length < 7 mm) dropped in response to hummingbird attacks 61 and 62% of the time (n = 90, Table 2). Leucauge venustra (Walckenaer) moved (25%), dropped (16%), or did not respond (42%) to hummingbird attacks (n = 12, Table 2). Adult N. clavipes pumped (49%), dropped (7%), stilted (7%), reared (7%), or did not respond (28%) to hummingbird attacks (n = 14, Table 2). 72 THE JOURNAL OF ARACHNOLOGY Table 2. -The responses of orb-weaving spiders to predation by the hummingbird Phaethornis superciliosus . Unless otherwise indicated, all spiders were adult females. (2) = second instar spider, sex unknown; (E) = eggmass in the web. Behaviors Species No Response Prey Capture Pump Rear Stilt Move Drop Argiope argentata 2 5 2 Cyclosa caroli 5 1 1 Leucauge venustra 5 2 3 2 Nephila clavipes 4 7 1 1 1 Nephila clavipes (2) 4 6 3 20 Uloborus republicanus 18 33 Uloborus republicanus (E) 5 1 Pumping second instar Nephila clavipes were eaten readily, 80% of the time. Dropping and retreat afforded the best survival for small spiders (body length < 7 mm). Adult Leucauge venustra, Argiope argentata and N. clavipes survival was 100% regardless of behavior. This is apparently related to the large size of these spiders relative to the prey size range preferred by Phaethornis. Cyclosa caroli did not respond to 86% of the hummingbird attacks directed toward the debris-like stabilimentum (n = 7). This pattern is highly adaptive and characteristic of cryptic animals (Edmunds 1974, Robinson 1969). One out of three attacks involving C. caroli without debris was immediately successful. No attacks involving C. caroli with complete debris stabilimentum were successful and none was oriented toward the spider (n = 3); however, one attack to the Cyclosa with an incomplete stabilimentum (no debris below the spider) was successful. The differential response between Uloborus republicanus with and without egg- masses is of particular interest. Uloborus republicanus without eggmasses dropped in response to hummingbird predatory attacks. Uloborus republicanus with eggmasses did not respond to hummingbird attacks (Fisher exact test, P < 0.05). This differential response indicates that predator avoidance behaviors are modified to reflect the cost of capture. Dropping may not be adaptive for U. republicanus with an eggmass because movement attracts the hummingbird to the web. Once attracted to the web, there is a high probability that the hummingbird would eat the abandoned eggmass. The sequential production of eggmasses in Uloborus republicanus may involve a de- creasing fecundity or viability per eggmass, or, the energetic investment per eggmass may be sufficiently high to warrant eggmass protection (Anderson 1978). Nonetheless, it is adaptive for an individual uloborid to protect its investment in the eggmass. The lack of response to hummingbird attacks provides eggmass protection. CONCLUSION Dropping was most frequently employed by small spiders against hummingbirds. Drop- ping provides a rapid escape and protection by substrate vegetation. After dropping, the spider can return to the hub of the web via its dragline. Pumping was not an effective deterrent to hummingbird predation. HOFFM ASTER- ARANEID PREDATOR AVOIDANCE BEHAVIORS 73 Rapid lateral movements were the most frequent response to dorsal salticid attacks, Such movements tended to dislodge the salticid and throw it into the catching spiral where it was attacked. Pumping, used less frequently in response to salticid predation, also resulted in displacement of the salticid. Ventral attacks were countered by stilting or rapid lateral movements. Stilting interposes the hub between the salticid and the orb- weaver and frequently prevents successful captures. Data from hummingbird-uloborid encounters indicate that the behavioral response to predation can be modified by spiders with and without eggmasses. This variation is adaptive; by modifying the behavioral response to predation, the spider can protect its investment in the eggmass. ACKNOWLEDGMENTS I am deeply indebted to Michael and Barbara Robinson for their support and advice during my stay in Panama. I would like to thank George Anghr for mistnetting the hummingbirds used in this study. Susan Riechert and Carlos Varlerio provided helpful criticism of an earlier draft of the manuscript. G. B. Edwards identified the salticids. This study was supported by a SHRI summer research assistantship from Iowa State University and a Smithsonian visiting student fellowship. LITERATURE CITED Anderson, J. F. 1978. Energy content of spider eggs. Oecologia, 37:41-57. Edmunds, N. 1974. Defence in animals. Longman Inc., N.Y., 357 pp. Gardner, B. T. 1964. Hunger and sequential responses in the hunting behavior of salticid spiders. J. Comp. Physiol. Psych., 58:167-173. Robinson, M. H. 1969. Defenses against visually-hunting predators. Pages 225-259 in T. Dobzhansky, M. K. Hecht, and W. C. Steere (eds.). Evolutionary Biology, vol. 3. Appleton, Century, Crofts, N. Y., 309 pp. Robinson, M. H. and J. Olazarris. 1971. Units of behavior and complex sequences in the predatory behavior of Argiope argentata (Fabricius): (Araneae, Araneidae). Smithsonian Contrib. Zool., No. 65, 36 pp. Robinson, M. H. and B. Robinson. 1970. The stabilimentum of the orb-web spider Argiope argenta ta: an improbable defense against predators. Canadian Entomol., 102:641-655. Robinson, M. H. and C. Valerio. 1977. Attacks on large or heavily defended prey by tropical salticid spiders. Psyche, 84:1-10. Sokal, R. R. and F. J. Rohlf. 1969. Biometry. W. H. Freeman and Co., San Francisco., 776 pp. Tolbert, W. W. 1975. Predator avoidance behaviors and web defensive structures in the orb weavers Argiope aurantia Bind Argiope trifasciata. Psyche, 82:29-51. Young, A. M. 1971. Foraging for insects by a tropical hummingbird. Condor, 73:36-45. Manuscript received January 1980, revised April 1981. ' Tietjen, W. J. 1982. Influence of activity patterns on social organization of Mallos gregalis (Araneae, Dictynidae). J. Arachnol., 10:75-84. INFLUENCE OF ACTIVITY PATTERNS ON SOCIAL ORGANIZATION OF MALLOS GREGALIS (ARANEAE, DICTYNIDAE) William J. Tietjen1 Department of Biology Georgia College Milledgeville, GA 31061 USA ABSTRACT Direct observation of colony behavior indicated that individual Mallos gregalis spiders exhibited a change in behavior related to activity level and position in the web. High-activity M. gregalis were found on the surface of the web while low-activity animals were found in the web interior. Individuals move to the web surface in the evening and into the web interior at midday. The activity of the colony was recorded through photoelectric means and was circadian with the highest level of activity occurring at night, the lowest level occurring about midday. The circadian rhythmicity was maintained in isolated M. gregalis and activity was recorded by similar photoelectric means. The data indicate that isolates fluctuated between high and low activity levels over several days. I suggested that such changes in individual activity level allow each animal to partition its behavior between self-maintenance and colony-maintenance behaviors. In comparing the social structure of M. gregalis to other social arthro- pods, I indicated that the level of organization is more similar to that of tent caterpillars than to the social Hymenoptera. INTRODUCTION Only 33 group-living species have been described among the more than 32,000 known species of spiders (for review, see Burgess 1978, Kraft 1970, Kullman 1972, Shear 1970). One such group-living species is the dictynid Mallos gregalis (Burgess 1976, Diguet 1915). In the field more than 100,000 animals of both sexes and all stadia live together on one tree (Burgess 1979, Diguet 1909, Uetz, personal communication). Predation and feeding are communal with no cannibalism occurring among conspecifics (Witt et. al. 1978). Wilson (1971) called groupings of araneids similar in behavior to M. gregalis a “pre- social” level of organization. Burgess (1978) avoided the term “social” because M. gregalis lack a morphological or ethological caste system. Instead, he described the lifestyle as “communal cooperative”. One feature of communal organization is the circadian rhythms of the group, as measured through the synchrony of activity among group members and the organization of individual’s activity cycles (McBride 1976). Circadian activity rhythms have been observed in a variety of araneids (Cloudsley-Thompson 1978). The circadian activity rhythms of non-communal spiders were recently described in Araneus diadematus by Ramousse and Davis (1976) and in Agelenopsis aperta (Riechert and Tracy 1975). Krafft (1969) showed that activity in colonies of the communal spider Agelena consociata was Present address: The Lindenwood College, Department of Biology, Saint Charles, Missouri 63301. 76 THE JOURNAL OF ARACHNOLOGY circadian, peaking between 1900 and 2000 hr. Similarly, colony activity of Cyrtophora citricola is greatest at night with daytime movements restricted to prey capture and emergency web repair (Rystra 1979). No attempts were reported to record the activity of single animals for either communal spider. The present study is concerned with the activity rhythms and behavior of both isolated and grouped M. gregalis. The research was initiated to describe one aspect of the individual’s behavior as it contributes to the collective behavior of the colony. METHODS General Methods.— M. gregalis , from a small colony collected near Guadalajara, Mexico, by J. W. Burgess in 1974, were maintained in a climate-controlled room (T = 25 ± 5°C, RH = 90 ± 10%, 12.5 L: 1 1.5 D). One large colony occupied a web built on potted plants ( Philodendron spp.). Numerous smaller colonies built interconnected webs on the walls and among the exposed pipes near the ceiling. Colonies were fed houseflies ( Musca domestica ) at weekly intervals and daily were given a fine spray of water on the web- surface. Experimental animals were collected from the surface of the large web and from the interior of disassembled small colonies. Direct observations.— Approximately 200 individuals of both sexes and various age classes (mainly subadult and adult females; ca. 4-5 mm body length) were released within a large plexiglass cage (Fig. 1 ; 91 .0 Lx 47.0 W x 35.5 H cm). A narrow chamber located near one wall of the cage provided an area where a web was built and facilitated observa- tions of the spider’s movements within the interior of the web. The large cage was located in the climate-controlled room and the animals within were cared for in the same way as Fig. 1. -Plexiglass cage used for observation of colony behavior. Two screened vents (S, the second vent is located on the rear wall of the apparatus) provided ventilation. Plants (P) provided support for potential web sites. A narrow chamber (C) was formed along one wall by means of a plexiglass wedge (W) and allowed observation of animals in the web-interior. Plexiglass tunnels (T) allowed observations of animals in/on webs built among the plants. Hinged doors (D) allowed for access. TIETJEN- SOCIAL ORGANIZATION IN A COMMUNAL SPIDER 77 Observation periods were 15 min in length and occurred between 0900 and 1800 hr over a two-week period (12 hr of direct observation). The reduced light intensity and great number of highly active animals (see below) did not permit reliable observations outside of the above 9-hr period. Behaviors were recorded at 15-sec intervals and the position (inside or on the web-surface) was noted for each individual. Those animals in areas that were currently being used as prey-capture sites were scored as occupying the web-exterior. A second group of 36 adult female and two adult male M. gregalis were observed over a two-week period in a smaller plexiglass cage (20.5 x 17.5 x 18.0 cm; 15 hr observation). Lighting was natural for a May day in North Carolina and provided by a nearby window. Temperature was not controlled but was relatively constant at 23° C (the maximum temperature was about 4°C higher at 1600 hr than at the minimum temperature which occurred at 0500 hr). Observation periods were 15 min in length and randomly staggered over the two-week period to provide at least two observation periods per hour. Between 0800 and 1800 hr the numbers of moving and non-moving spiders, their behavior and positions in or on the web were stored in the memory of a microcomputer (programmed as an event recorded) and later transferred to magnetic tape. During the night-hours (1800-0700 hr) a 15 W lamp was adjusted to approximate a moonlit night. However, due to the low level of light available and the large number of active animals, only the position of each animal and level of activity were recorded during the night -hours. In addition to the above systematic observations, 40-50 hr were spent in less struc- tured observation of the large colony on the table. Most of these observations occurred between 0900 and 2000 hr and only the behaviors of animals on the surface of the web were visible. Photoelectric recordings.— The activity levels of 50 isolated adult female M. gregalis were recorded in activity chambers (Fig. 2). Females were contained in a 40 mm (id) x 170 mm (1) tygon tube forming a toroidal activity chamber (ca. 50 mm dia). The passing of a spider between the phototransistor and light source was recorded on one of 20 channels of an Esterline-Angus event recorder. Further information on this apparatus is Fig. 2-Photoelectric recording methods. A: Apparatus to measure activity of isolated Mallos gregalis. Tygon tubing (T), held end-to-end by a coupling tube (C) contained the animals. Movement of spiders between the phototransistor (P) and biasing lamp (L) was recorded on one channel of an event recorder (ER). Colony activity (B) used a photocell matrix (PM). Movement of spiders on the web (W) was recorded as in A. 78 THE JOURNAL OF ARACHNOLOGY available from the author. Animals were provided daily with a drop of water but were not fed during the experiments. Such periods of deprivation are well within the survival limits recorded for a variety of spiders which range from several months to nearly a year (Anderson 1974, Miyashita 1968, Turnbull 1962, 1965, Witt 1963). Photoperiod was approximately 10 L: 13 D and the maximum temperature (0900-1000 hr) was 27° C which decreased to about 22.5° C at 0600 hr. Humidity was not controlled and varied between 70 and 95%. Following a 24 hr acclimatization period for the females, experiments ran contin- uously for 9 days. Only those spiders surviving the entire experimental period and an additional day following the experiments were used in the analysis of data (33 animals; most deaths were due to drowning in the water droplet). The activities of 1 2 isolated adult female M. gregalis were recorded in a similar manner to the above group, except that illumination was continuous over an 11 -day period, starting 3 days prior to the 9-day recording period. In these experiments deaths due to drowning were eliminated. One female died from undetermined causes. The activity of a small colony (74 adult females) was recorded by photoelectric methods in a 20.5 x 17.5 x 18.0 cm plexiglass cage (fig. 2). Twelve phototransistors were arrayed in a 2 x 6 matrix and positioned to record activity accurring mainly on the surface of the web. A 15-W incandescent lamp located 1.5 m from the photosensors provided the biasing source and approximated a bright moonlit night. Movements of the animals were recorded on an event recorder. Krafft (1969), in a similar manner, recorded activity in a small colony (N = 5 individuals) of the colonial spider Agelena consociata. Before recordings began, a 7-day period was provided to allow the spiders to build a web within the cage. Twelve houseflies (M. domestica) were fed to the spiders on the last of the 7 days and activity was continuously recorded beginning on the following day for a 12-day period. Environmental conditions were similar to those provided for the isolates. Animals were daily given a fine spray of water on the web-surface. Statistical analyses were performed according to the methods of Batschelet (1965), Conover (1971) and Sokal and Rohlf (1969). All means are accompanied by their stan- dard errors. RESULTS Observations of colony behavior.— Casual observations suggested that, although prey were occasionally pulled into the web-interior and fed upon, communal feeding occurred only on the web-surface. Fecal material was deposited mainly at the edge of the web- surface as evidenced by a ring of droppings on the table and floor surfaces surrounding the web. Few animals were observed on the web-surface during the midday hours, while in the evening or early morning hours the surface was covered with hundreds of active M. gregalis. Since it was not possible to determine the position of all animals during an observation period in the large plexiglass cage, movements between the web-surface and interior were not analyzed. Only the types of behaviors and the active animals’ positions are included in this data set and these results are pooled with the data from the small plexiglass cage. In the small cage, the percentage of animals moving per 1 5 sec on the web-surface (X = 14.7 ± 0.4%) greatly exceeded movement in the web-interior throughout the day (X = 1 .8 ± 0.1%; Mann Whitney test, p < 0.001). Movements in the web-interior and on the web-surface were not correlated (Spearman’s rho,p =+0.41 1, p <0.10). The majority of TIETJEN- SOCIAL ORGANIZATION IN A COMMUNAL SPIDER 79 animals moved to the web-surface during the evening and returned to the web-interior by midday (Fig. 3; Raleigh test, p < 0.01). Many behaviors were mainly observed in one web-location. Male courtship behavior (N = 1 5) was observed only in the web-interior, as was mating (N = 5). All egg sacs were deposited in the interior of the web. The incidence of cribellate silk deposition on the web-surface (N = 207/15-sec intervals) greatly exceeded deposition in the web-interior (N = 59/15-sec intervals; x2 test adjusted for aminal position, p < 0.001). These data were supported by disassembly of webs: silk in the web-interior was compact and non-adhesive while silk on the exterior was adhesive. Colony activity peaked at 0351 hr ± 0455 and the lowest activity occurred about midday (Raleigh test; p < 0.01 as measured by the photoelectric method and pooled over the experimental period, Fig. 4). These results are consistent with those obtained through direct observations and actographic methods (Tietjen, unpublished data). The mean total events per day was 57.0 ± 8.1 (one event is the movement of an animal between a phototransistor and the source of light) and activity did not vary among days (Wilcoxon test, p > 0.50). No trends were observed in the activity level over the experimental period (Spearman’s rho, p = +0.14, p > 0.10). Observations of isolates.— Preliminary observations indicated that the spiders moved freely through the recording area and were not attracted to any one area of the apparatus (including the low-level light used to bias the phototransistors). Those animals recorded as inactive were usually quiescent and did not exhibit behaviors such as silk-deposition outside the field of view of the phototransistor. The mean movement per animal was 15.7 ± 2.3 events per day. The mean difference between the maximum and minimum daily activity level per animal was 34.0 ± 12.4 events; thus, during days of decreased activity animals typically exhibited little or no movement while days of high activity were characterized by extensive movement. Eight animals exhibited no significant differences in activity over the 9-day period (x2 test, p > 0.05) while the remaining 25 exhibited a significant difference among days (p < 0.05). TIME Fig. 3. -Circadian movement between web-areas. The graph indicates the percent animals observed on the web-surface over a 24-hr period (pooled for 2-hr intervals). 80 THE JOURNAL OF ARACHNOLOGY The difference in total daily activity compared among spiders varied considerably (x2 test, p < 0.0001). Only three of the 33 spiders exhibited a positive trend in daily activity level over the 9-day period while no spiders exhibited a negative trend (Spearman’s rho, p < 0.05 for positive correlation with time). Figure 5 indicates the daily activity level of each animal over the 9-day period. Note that most spiders shift between high and low activity levels over several days. The total numbers of events per hour were pooled among isolated spiders thus treating the isolates as if they occupied a single colony. Data indicate an activity peak at 0253 ± 0429 hr and a decrease in activity about midday (Fig. 4; Raleigh test, p < 0.01). No differences were observed between the activity peaks for the pooled isolates and the colony (F test, p < 0.10). The pooled activity of all spiders did not vary among days (x2 = 519.6 ± 68.2 events per day: Kruskal-Wallis test, p> 0.05). No correlation was observed between the total events per day and time since recording began (Spearman’s rho, p = 0.35, p>0.10). Isolates tested under constant-light conditions exhibited a decrease in activity com- pared to isolates under normal lighting conditions (x2 = 6.9 ± 0.5 events per day; Mann Whitney test, p < 0.0001). The daily activity level was too low to test on a day-to-day basis and the data were therefore pooled for further analysis. The greatest activity level occurred at 0.617 hr ± 0428. This was not significantly different from the activity cycles observed for the colony or isolates under normal lighting conditions (F test , p > 0.05). DISCUSSION Colony behavior.— The present data indicate that the collective activity levels of in- dividual M. gregalis in a colony are highest at night and lowest at midday. Krafft (1969) described a similar circadian activity rhythm in the communal spider A. consociata and indicated that the timing of the cycle was influenced by photoperiod. My data, however, offer conflicting evidence. Although the constant-light condition affected the total daily activity level of the isolates, the expected shift in time for peak activity was not observed. This may be due to the necessity of pooling the data over the experimental period or other cues, such as temperature, may also be involved as is the case in solitary spiders (Witt 1963). 60 TIME Fig. 4.-Circadian activity of isolated and groups Mallos gregalis. Activity of a colony pooled over the experimental period is indicated by the strippled bars. The mean activity level of isolated animals pooled among animals is indicated by the hatched bars. TIETJEN- SOCIAL ORGANIZATION IN A COMMUNAL SPIDER 81 Direct observations indicate that activity level is higher on the web-surface than in the web-interior throughout the day. The spiders did, however, exhibit a circadian rhyth- micity in movement between the web areas; occupying the web-surface mainly at night and the web-interior during the day. Animals in the web-interior were typically inactive and the lack of a positive correlation between activity levels in the two areas suggests that, as spiders in the web-interior become active, they move to the web-surface. In nature, spiders may move to the web-interior to avoid the heat of the noonday sun. Such temperature-dependent movements have been described in the non-communal spider Agelenopsis aperta (Richert and Tracy 1975). In addition, by restricting periods of activity in the night hours, M. gregalis may effectively avoid visual predators (e.g. birds) as suggested by Rypstra (1979) for the communal spider Cyrtophora citricola. Factors other than activity may have their probability patterns affected by the ani- mal’s position on the web. Silk-deposition, communal-feeding, prey capture and fecal- elimination occurred mainly on the web-surface. In contrast to the report of Jackson and Smith (1978), males exhibited courtship behavior only in the web-interior. All matings and deposition of egg sacs similarly occurred only in the web-interior. The web therefore appears to assist in organizing and coordinating the various behaviors of colony members. Alternatively, the animals may organize their behavior spatially and build the web accord- ingly, however, such an explanation would be applicable only under very restricted con- ditions. Tietjen (1981) indicated that the webs built by colonies in small containers (petri dishes) tended to be densest near the rim of the dish while colonies built under less confined conditions were highly variable in structure (Hollar, personal communication, Tietjen, unpublished data). Effects of isolation. -Among the Araneae, there are few studies regarding the effect of group-size on the behavior of the individual. Burch (1977, 1979) investigated the effect of isolation during the early communal life of Araneus diadematus, and showed that, although isolates built functional webs, they matured more slowly than spiderlings raised communally. Among the social spiders, isolated mature female Stegodyphus sarasinorum 1 a> 10 _Q £ Z 1 20 c < 30 Fib. 5. -Daily activity of isolates. A graphical representation of the day-to-day activity of 33 animals over the 9-day experimental period is presented. Days of an individual’s increased activity (number of events per day greater than Y + 1SE) are indicated by the shaded bars; days of low activity are indicated by unshaded bars. 82 THE JOURNAL OF ARACHNOLOGY are less active than conspecifics living communally (Tietjen, unpublished data) and fewer A. consociata survive when isolated from nestmates (Krafft 1969). Mallos gregalis , on the other hand, develop and survive normally when isolated from the colony (Jackson 1979). In addition, individuals often leave the parent colony and produce functional webs. The present data suggest that the circadian activity rhythms of individuals were not affected by isolation since the circadian activity of the colony was similar to the pooled activity of isolates (Fig. 4). Individual activity patterns and the effect on colony organization.— M. gregalis lacks a distinct caste system and all members of the colony perform a variety of tasks related to web-maintenance, web-building, feeding, prey capture and reproduction (Burgess 1976). Coordination of some or all of these tasks may depend on individual activity levels. Individual M. gregalis exhibit fluctuations in activity over several days (Fig. 5) and the level of activity affected the spider’s position on the web. In addition, the likelihood of certain behaviors occurring was related to web-position. These data suggest that individual activity levels coordinate colony-maintenance vs. self-maintenance behaviors in the com- munal web. Animals of low activity would be expected to remain in the web-interior; involved mainly in self-maintenance behaviors such as reproduction or feeding on flies drawn into the interior (although the majority of feeding occurs on the catching surface). Animals of high activity would be found on the exterior of the web taking part in colony-maintenance behaviors such as prey-capture and silk-deposition. One might expect that such a system might lead to “cheaters” in which an animal spends most of its time in self-maintenance dedicating little or none of its time budget to colony-maintenance be- haviors. Physiological and/or behavioral restraints such as the need to empty silk glands or to move to the surface for feeding may inhibit individuals from becoming cheaters. In Fig. 6- Individual activity cycles summate and result in constant daily activity levels for the colony. A) The probability of movement (m) is affected by the circadian cycle (dotted line) and the individual’s cycle (solid line). In this example, high activity occurs on days 1, 4 and 5 (hatched areas). The individual cycles and periods of high daily activity are shown for two additional animals (B, C). TIETJEN- SOCIAL ORGANIZATION IN A COMMUNAL SPIDER 83 addition, for long-term nest sites one might expect a high degree of relatedness among nest mates allowing kin selection to occur. Such interpretations must, however, await a genetic analysis of a number of natural populations. The factors which control individual daily activity were not determined. It is unlikely, however, that the level of hunger in these experiments affected total daily activity (at least over the experimental period), as animals were not fed and few trends were observed in total daily activity. Likewise temperature, humidity and photoperiod have little effect on an animal’s total daily activity since isolates were tested concurrently and the peaks of high daily activity were out of phase among animals (Fig. 5); these factors may be important in nature as they affect prey density. Endogenous conditions such as sexual tone (Crane 1949) may affect daily activity. Individuals of high sexual tone, for example, would be expected to exhibit low activity and be found mainly in the web-interior. When the total daily activity among animals was pooled and analyzed day-to-day, the total daily activity of the “statistical colony” did not change over the experimental period even though each animal’s contribution to the activity changed over several days (Fig. 5). Likewise, the total daily activity of the colony recorded by photoelectric methods did not change over the experimental period. Figure 6 represents a model of the summation of individual activity cycles and their effect on colony activity. Two cycles are apparent: 1) A circadian cycle shared by all members of the group, cued perhaps by a combination of photoperiod and temperature and 2) an individual cycle, cued by endo- genous factors and out of phase among animals. High levels of individual activity are possible only when both cycles are at a peak. Low activity occurs under all other circum- stances. With many individuals in a colony, each with their own personal activity cycle, the end result is a total daily colony activity which does not change on a day-to-day basis even though the active participation of individuals changes constantly. In the tent caterpillar larve ( Malocosoma pluviale) an analogous situation exists. In- dividual differences in activity affect the behavior and development of the colony. Larve of high activity tend to emigrate and become leaders of the group while those of low activity are the followers. Individual activity levels does not, however, change from day- to-day but extend throughout adult life. The proportion of active and inactive larve in the colony affects the behavior of the entire group; controlling, for example, the frequency of future infestations and the propensity of the colony to defoliate new areas of the host tree (Wellington 1957). Among the social Hymenoptera extreme minor-major caste polymorphism, as seen in some species of ants and termites, is accompanied by behavioral specializations which divide the work of the colony according to caste (Wilson 1971). Among the social bees, individual workers have a limited behavioral repertoire; however, individual and age polyethisms allow for efficient foraging among a variety of flower-types (Heinrich 1976). According to the data presented, the organization of colonies of M. gregalis is therefore more similar to that of tent caterpillers than to the organization seen in the social Hymenoptera. ACKNOWLEDGMENTS I am grateful to Dr. P. N. Witt for many helpful discussions during the course of this research and the preparation of the manuscript. Drs. S. E. Riechert and G. W. Uetz provided helpful comments in the review of the manuscript. The research was supported by National Science Foundation grant BNS 75-09915 to P. N. Witt and W. J. Tietjen and was conducted in the laboratories of the North Carolina Department of Mental Health. 84 THE JOURNAL OF ARACHNOLOGY REFERENCES CITED Anderson, J. F. 1974. Repsonses to starvation in the spiders Lycosa lenta Hentz and Filistata hiber- nalis (Hentz). Ecology, 55:576-585. Batshelet, R. 1965. Statistical methods for the analysis of problems in animal orientation and certain biological rhythms. Washington, D. C.: Amer. Inst. Bio. Sci., 57 pp. Burch, T. L. 1977. Early communal life of the solitary web spider (Araneus diadematus Cl.) Raleigh: Ph.D. Thesis N.C. State Univ., 6-77. Burch, T. L. 1979. The importance of communal experience to survival for spiderlings of Araneus diadematus (Araneae: Araneidae). J. Arachnol. 7:1-18. Burgess, J. W. 1976. Social Spiders. Scient. Am., 234:100-106. Burgess, J. W. 1978. Social behavior in group-living spider species. Symp. Zool. Soc. London, 42:69-78. Burgess, J. W. 1979. Mallos gregalis- behavioral adaptations for living in Mexico. Raleigh: Ph.D. Thesis N. C. State Univ., 3-20. Cloudsley-Thompson, J. L. 1978. Biological clocks in Arachnida. Bull. Br. Arachnol. Soc., 4:184-191. Conover, W. J. 1971. Practical nonparametric statistics. New York: John Wiley and Sons, Inc., 492 pp. Crane, J. 1949. Comparative biology of salticid spiders at Rancho Grande, Venezuela. Part IV. An analysis of display. Zoologica, 34:159-215. Diguet, L. 1909. Sur faraignees mosquero- Compte Rendu Acad. Sci. Paris, 148:735-736. Diguet, L. 1915. Nouvelles observations sur le mosquero ou nid d’araignees sociales. Bull. Soc. nat. acclim. France, 62:240-249. Heinrich, B. 1976. Bumblebee foraging and the economics of sociality. Amer. Sci., 64:384-395. Jackson, R. R. 1979. Predatory behavior of the social spider Mallos gregalis: is it cooperative? Insectes Soc., 26:300-312. Jackson, R. R. and S. E. Smith. 1978. Aggregations of Mallos and Dictyna (Araneae, Dictynidae): Population characteristics. Psyche, 85:65-80. Krafft, B. 1969. Various aspects of the biology of Agelena consociata Denis when bred in the labora- tory. Amer. Zool., 9:201-210. Krafft, B. 1970. Contribution a la biologie et a l’ethologie d’ Agelena consociata Denis (Araignee social du Gabon). Biol. Gabon, 6:199-301. Kullman, E. J. 1972. Evolution of social behavior in spiders (Araneae: Eresidae and Theridiidae). Amer. Zool., 12:419-426. McBride, G. 1976. The study of social organization. Behavior, 59:96-115. Miyashita, K. 1968. Quantitative feeding biology of Lycosa T-insignata Boes. et Str. (Araneae: Lycosidae). Bull. Nat. Inst. Agric. Sci., Ser. C., 22:329-334. Ramousse, R. and F. Davis III. 1976. Web-building time for a spider: preliminary application of ultrasonic detection. Physiol. Behav., 17:997-1000. Riechert, S. E. and C. R. Tracy. 1975. Thermal balance and prey availability: basis for a model relating web-site characteristics to spider reproductive success. Ecology, 56:265-284. Rypstra, A. L. 1979. Foraging flocks of spiders. Behav. Ecol. Sociobiol., 5:291-300. Shear, W. A. 1970. The evolution of social phenomena in spiders. Bull. Br. Arach. Soc., 1:65-76. Sokal, R. R. and F. J. Rohlf. 1969. Biometry. San Francisco: W. H. Freeman, 766 pp. Tietjen, W. J. 1981. Sanitary behavior by the social spider Mallos gregalis (Dictynidae): Distribution of excreta as related to web density and animal movements. Psyche (in press). Turnbull, A. L. 1962. Quantitative studies of the food of Linyphia triangularis Clerck (Araneae: Linphilidae). Canadian Ent., 94:1233-1349. Turnbull, A. L. 1965. Effects of prey abundance on the development of the spider Agelenopsis potteri (Blackwall) (Araneae: Agelenidae). Canadian Ent., 97:141-147. Wellington, W. G. 1957. Individual differences as a factor in population dynamics: the development of a problem. Canadian J. Zool., 35:293-323. Wilson, E. O. 1971. The Insect Societies. Cambridge, Mass.: Harvard Univ. Press, 548 pp. Witt. P. N. 1963. Environment in relation to behavior of spiders. Arch. Environ. Health, 7:4-12. Witt, P. N., M. B. Scarboro, and D. B. Peakall. 1978. Comparative feeding data in three spider species of different sociality: Araneus diadematus Cl., Mallos trivittatus (Banks) and Mallos gregalis (Simon). Symp. Zool. Soc. London, 42:89-97. Manuscript received August 1980, revised February 1981. The Journal of Arachnology 10:85 RESEARCH NOTES DIURNALISM IN PARABUTHUS VILLOSUS (PETERS) (SCORPIONES, BUTHIDAE) It is well known most scorpions are strictly nocturnal in their activities although definite, restricted cases of diurnal behavior have been recorded. Williams (1971, 1980) described the phenomenon in the vaejovid Vaefovis littoralis Williams, and Roth and Brown (1976) found it in another species, probably Vaejovis gertschi Williams (Williams pers. comm.), both occurring in Baja California. Toye (1970) found the scorpionid Pandinus imperator Koch occasionally active by day in Nigerian forests. In Namibia, Lamoral (1979) has observed diurnalism in the buthids Buthotus conspersus (Thorell), Parabuthus kraepelini Werner, Parabuthus stridulus Hewitt, and in the scorpionid Opisthophthalmus carinatus (Peters). The diurnal behavior of some of these is probably fortuitous since they are predominantly nocturnal. The present note describes in further detail the partly diurnal behavior (Newlands 1974) of Parabuthus villosus (Peters). This species is one of the largest buthid scorpions known and is widespread in Namibia. Its distribution stretches roughly from the northern Cape Province, South Africa, to northern Damaraland. From west to east it ranges from the coast to as far inland as longitude 19°30\ It is found in the extremely arid gravel plains of the Namib desert though is apparently absent in the sand dune systems of the Namib and Kalahari deserts. Diurnal sightings of P. villosus (predominantly adult males and females) were made by various individuals, including the author, and are listed below. Most were made at Gobabeb in the central Namib where a permanent research station exists. The localities cover most of the range of the species. Gobabeb 0830 hrs; Valencia 42, Windhoek district 0910; Fish River Canyon 1000; Bethanie 1100; Gobabeb 1300; (1 July 1980, hot); Windhoek 1300-1400; Namuskluft 88, Liideritz district 1400; Gobabeb 1445; (6 June 1980, feeding on beetle); Gobabeb 1400-1500; Gobabeb 1500 (warm); Gobabeb 1500 (April, very hot); Uis 1600; Hentiesbay 1600 (foggy); Gobabeb 1600 (December); Gobabeb 1630; Sandmund 270, Keetmanshoop district 1630 (cool); Obib mountains 1600-1700 (fairly hot); Omaruru river 1600-1700; Gai-As Fountain, Damaraland 1600-1700; Gobabeb 1715; Fish River Canyon 1745 (2 May 1979); Uis 1800 (twilight). It is thus evident that P. villosus can be active throughout the day, mostly during cool, overcast or twilight conditions although definite observations to the contrary exist. In addition, the vaejovids already mentioned by Williams (1971 , 1980) and Roth and Brown (1976), and to an extent P. villosus, are active during the hottest part of the day. Compared to that of some closely related scorpions in Namibia, diurnal behavior in P. villosus is distinctly atypical. Neither Parabuthus granulatus (Hemprich and Ehrenberg), a widespread burrower, Parabuthus raudus (Simon), also a burrower, in the Kalahari Sand System, nor the rarer Parabuthus schlecteri Purcell, occurring under stones in southern Namibia, have been seen to be active by day. In addition, and in spite of a The Journal of Arachnology 10:86 higher population density of humans within its range, no daytime activity has been reported for the common Parabuthus transvaalicus Purcell. This species is closely related to P. villosus and also ecologically equivalent to it.P. transvaalicus is found under stones in arid areas of the northern Transvaal, eastern Zimbabwe and southwestern Botswana, all roughly similar in rainfall and vegetation to the Namibian habitat of P. villosus. The reasons underlying diurnal activity in scorpions are poorly understood, though may be partly accounted for by being an adaptation for better exploitation of prey resources. Thus it is perhaps significant that both P. villosus and the vaejovids already mentioned have been reported to feed by day. The former has been seen catching beetles, at Gobabeb in the Namib desert (Marinaki, pers. comm.). Similarly, the vaejovids de- scribed by Williams (1971) and Roth and Brown (1976) were reported to stalk insects and eat isopods, respectively. Certainly in the case of P. villosus , on the gravel plains of the central Namib at Gobabeb, more beetles, a major food source for scorpions, are active by day than by night (Seely, pers. comm., Koch 1961). Thus, the diurnal activity seen in this scorpion may have evolved to take advantage of such daytime food supplies. Some arachnids other than the scorpions described above show a similarly unexpected partial diurnal activity, hexisopodid solifugids of the genera Chelypus and Hexisopus (Lawrence, pers. comm., Lamoral 1973) being examples. These “mole” solifugids are highly specialized, adapted to a psammophile, fossorial way of life and when diurnal, are known to be active during the hottest part of the day (Lawrence, pers. comm.). In contrast to typical diurnal solifugids such as Solpuga sp., however, they lack the charac- teristically rapid locomotion and dense pubescence of the latter. I thank Mr. G. Newlands and Dr. 0. F. Francke for helpful comments and Dr. S. C. Williams and Mr. V. D. Roth for their interest in diurnalism in scorpions. The loan of the specimen of Vaejovis littoralis by Dr. D. Kavanaugh of the California Academy of Sciences is acknowledged, as is the interest of Dr. R. F. Lawrence and Dr. M. K. Seely. LITERATURE CITED Koch, C. L. 1961. Some aspects of abundant life in the vegetationless sand of the Namib Desert dunes. J. Southwest Africa Sci. Soc., 15:9-33. Lamoral, B. H. 1973. The Arachnid fauna of the Kalahari Gemsbok National Park. Part 1. A revision of the species of “Mole Solifuges” of the genus Chelypus Purcell, 1901 (family Hexisopodidae). Koedoe, 16:83-102. Lamoral, B. H. 1979. The Scorpions of Namibia (Arachnida: Scorpionida). Ann. Natal Mus., 23:497-784. Newlands. G. 1974. The venom-squirting ability of Parabuthus scorpions (Arachnida: Buthidae). South African J. Med. Sci., 39:175-178. Roth, V. D. and W. L. Brown. 1976. Other intertidal air-breathing arthropods. In L. Cheng (ed.), Marine insects. North Holland Publishing Company, Amsterdam, 581 pp. Toye, S. A. 1970. Some aspects of the biology of two common species of Nigerian scorpions. J: Zool. London, 162:1-9. Williams, S. C. 1971. In search of scorpions. Pacific Discovery, 24:1-10. Williams, S. C. 1980. Scorpions of Baja California, Mexico and Adjacent Islands. Occasional Papers, California Acad. Sci., 135:1-127. Alexis Harington, Department of Zoology, University of the Witwatersrand, 1 Jan Smuts Avenue, Johannesburg 2001 , South Africa. Manuscript received January 1981, revised April 1981. The Journal of Arachnology 10:87 A NEW RHINOCHERNES FROM ECUADOR (PSEUDOSCORPIONIDA, CHERNETIDAE) Among the pseudoscorpions collected by N. P. Ashmole in and around Los Tayos Caves, Ecuador, is a specimen which deserves to be described at this time. A female, it apparently represents the genus Rhinochernes Beier (1954), which has hitherto been known only from males of a single species from Peru. Rhinochernes ashmolei , new species Figs. 1-4 Material.— Holotype female (WM 4694.01001) taken from vegetation near Los Tayos Caves, Cordillera el Condor, Ecuador (3°04'S, 78°15'W) on 22 July 1976 (N. P. Ash- mole). The specimen is in the Florida State Collection of Arthropods, Gainesville. Diagnosis.— Generally similar to Rhinochernes granulatus Beier but smaller and with less attenuated appendages; palpal femur 0.43 mm long, 1/w ratio 2.05, and chela 0.84 mm long, 1/w ratio 2.3. Description of female (male unknown).— Color generally light brown. Carapace a little longer than broad, with 2 distinct transverse furrows and a small but distinct median elevation near posterior edge; surface granulate; 2 smooth eyespots; about 36 stout, Figs. 1-4 .-Rhinochernes ashmolei, new species: 1, genital opercula of female; 2, spermathecae; 3, dorsal view of right palp; 4, lateral view of left chela. The Journal of Arachnology 10:88 terminally denticulate setae, with 4 at anterior and 7 at posterior margin. Tergites 1-10 and sternites 4-10 divided; surface of tergites granulate, of sternites smooth; pleural membranes longitudinally rugose and papillose . Tergal chaetotaxy 9:12:12:11:12:11:12: 12:12:8:8:2; sternal chaetotaxy 17:(2)10(2):(1)6(1): 12: 15: 17: 13: 1 1 :T1TT1T:2; tergal setae, stout, terminally denticulate, sternal setae mostly acuminate, Genital opercula as shown in Fig. 1 ; spermathecae consisting of 2 long, irregularly expanded tubules (Fig. 2). Chelicera 2/5 as long as carapace; hand with 5 setae, sb laterally denticulate, b and es acuminate; flagellum of 3 long, slender setae, the distal one serrate; galea large, with 3 branched rami; serrula exterior with 17 blades. Palp rather heavy (Fig. 3); femur 2.05, tibia 1.9, and chela (without pedicel) 2.3 times as long as broad; hand (without pedicel) 1.1 times as long as deep; movable finger 0.96 as long as hand. All surfaces granulate; setae short, stout, terminally denticulate; many set in raised areoles, especially on medial sides of femur, tibia and chela. Fixed chelal finger with 29 and movable finger with 33 well developed, cusped marginal teeth; each finger with 1 internal and 7 or 8 external accessory teeth. Each finger with a venom duct, but that of fixed finger much reduced. Trichobothria on chela as shown in Fig. 4; those of fixed situated as in R. granulatus , but on movable finger st much closer to t than to sb. Legs relatively slender; leg IV with entire femur 3.1, tibia 3.6, and tarsus 4.5 times as long as deep. Tarsus of leg IV with a long tactile seta just distal of middle. Measurements (mm).— Body length 2.0. Carapace length 0.635. Chelicera 0.26 by 0.13. Palpal trochanter 0.32 by 0.19; femur 0.43 by 0.21; tibia 0.44 by 0.23; chela (without pedicel) 0.84 by 0.37; hand (without pedicel) 0.45 by 0.41; pedicel about 0.07 long; movable finger 0.43 long. Leg IV: entire femur 0.50 by 0.16; tibia 0.37 by 0.10; tarsus 0.34 by 0.07. Etymology.— The species is named for N. Philip Ashmole who collected the specimen. Remarks.— Except for size and proportions, this form is rather similar to R. granulatus. However, it must be noted that it differs in two details from the description given by Beier. On the movable chelal finger trichobothrium st lies very close to t rather than “in der Mitte zwischen sb and f” (p. 7); and the tactile seta on the tarsus of leg IV is situated near the middle of the segment rather than “halbwegs zwischen der Mitte und dem Ende des Gliedes” (p. 9). Also, it should be pointed out that the carapace has a distinct median eminence near the posterior margin reminiscent of that (“keel”) to be seen in most species of Parachernes (see Muchmore and Alteri 1974). It seems best for now to place the species in Rhinochernes , and try to resolve the differences later when more material becomes available. I am greatly indebted to Dr. Ashmole for sending the specimens and to C. H. Alteri for preparing the illustrations. LITERATURE CITED Beier, M. 1954. Pseudoscorpionidea. Beitr. Fauna Perus, 4:1-12. Muchmore, W. B. and C. H. Alteri. 1974. The genus Parachernes (Pseudo scorpionida, Chernetidae) in the United States, with descriptions of new species. Trans. Amer. Entomol. Soc., 99:477-506. William B. Muchmore, Department of Biology, University of Rochester, Rochester, New York 14627. Manuscript received July 1980, accepted September 1980. The Journal of Arachnology 10:89 COMMENTS ON SOME LEIOBUNUM SPECIES OF THE U S A. (OPILIONES; PALPATORES, LEIOBUNIDAE) Since the revision of the genus Leiobunum by Davis (1934), ten new species have been described from the United States. Goodnight and Goodnight (1943, 1945) described L. oregonense, L. trimaculatum, and L. gordonv, Roewer (1952, 1957) described L. caver- narum, L. davisi, L. zimmermani, and L. supracheliceralis', Edgar (1962) described L. lineatum; and McGhee (1977) described L. bracchiolum and L. holtae. Two of these species have been placed in synonomy by McGhee (1977) and Shear (1980). The remain- ing eight species are predominantly eastern in distribution; only L. oregonense and L. supracheliceralis are known from the western half of the United States. The latter two are now known to be synonyms of previously described species. Examination of the holotype of L. oregonense reveals it is conspecific with L. paessleri Roewer, 1910. Leiobunum paessleri is a relatively common species found from central California along the coastal states to central Alaska. The holotype of L . oregonense was collected at Rain Rock, Lane Co., Oregon by Borys Malkin and is housed at the American Museum of Natural History (AMNH). The two male paratypes (reported as male and female) of L . oregonense were also examined. Both were found to be L. depressum Davis, 1934. I have also examined the male holotype of L. depressum (coll. AMNH) and find it slightly smaller and darker in color than either of the L. oregonense paratypes, but no other differences could be found. The paratypes of L. oregonense were reported to have been collected at Car Canyon, California, but labels with the specimens indicate other- wise: “Carr Cyn, Huachuca Mts. Ariz VIII-9-40 (E. S. Ross)”. The paratypes are in the collection of the California Academy of Sciences. Leiobunum depressum is known, thus far, only from Arizona, New Mexico, and Utah. Throughout most of the range of L. depressum, L. townsendi Banks will be en- countered. Both species are very similar, particularly males, and are often hard to separate on the basis of external characters. In his key, Davis clearly separated these two based on the white leg bands present in L. townsendi. However, most L. townsendi from western New Mexico and Arizona have leg bands which are indistinct or absent. The median longitudinal keel on the penis shaft of L. depressum is now known to be present in several other species, including L. townsendi. My examinations reveal only two external characters that might separate these two species when leg bands are absent. Generally L. townsendi will have a white ring around the eyes, and the coxae will be smooth except for the marginal rows of denticles. The area around the eyes of L. depressum is always dark, and the coxae are covered with low rounded tubercles plus the marginal rows of denticles. The best method to separate L. townsendi from L. depressum is examination of the genitalia. The distal end of the penis shaft and the proximal portion of the glans are extremely wide (Figs. 1, 3) in L. townsendi. In L. depressum the glans and shaft are compressed at their junctions (Figs. 2, 4). The seminal receptacles of L. townsendi and L. depressum are quite dissimilar (Figs. 5, 6). The male holotype and two male paratypes of L. supracheliceralis were reported to have been collected in “Texas” and are in the collection of the Senckenberg Natur- Museum. Examination of this series reveals that all are L. flavum Banks, 1894. The enormous supracheliceral lamellae illustrated by Roewer (1957: Fig. 15) are exaggerated. The type specimens appear quite normal for L. flavum. As L. flavum is presently known only from the eastern portion of Texas, Roewer’s specimens probably are from that region. The Journal of Arachnology 10:90 Figs. 1-6. -Male and female genitalia: 1-2, Dorsal aspect of penes: 1, L. townsendi; 2, L. depression ; 3-4, Ventral aspect of distal end of penes: 3, L. townsendi’, 4 ,L. depression; 5-6, Seminal receptacle: 5, L. townsendi ; 6, L. depressum. Scale line = 0.22 for penes, 0.05mm for seminal receptacles. I would like to thank Manfred Grasshoff, David H. Kavanaugh, Vincent F. Lee, and Norman I. Platnick and their institutions for allowing me to examine the type series. LITERATURE CITED Davis, N. W. 1934. A revision of the genus Leiobunum (Opiliones) of the United States. Amer. Midland Nat., 15(6):662-705. Edgar, A. L. 1962. A new phalangid (Opiliones) from Michigan, Leiobunum lineatum sp. nov. Trans. Amer. Micros. Soc., 81:146-149. Goodnight, C. J. and M. L. Goodnight. 1943. New and little known phalangids from the United States. Amer. Midland Nat., 29(3):643-656. Goodnight, C. J. and M. L. Goodnight. 1945. Phalangida from the United States. J. New York Entomol. Soc., 53:239-245. McGhee, C. R. 1977. The politum group (bulbate species) of Leiobunum (Arachnida: Phalangida: Phalangidae) of North America. J. Arachnol., 3:151-163. Roewer, C. F. 1952. Einige Phalangiiden aus dem Vereinigten Staaten von Nord-Amerika. Zool. Anz., 1 49(1 1/12): 267-27 3. Roewer, C. F. 1957. Uber Oligolophinae, Caddoinae, Sclerosomatinae, Leiobuninae, Neopilioninae und Leptobuninae (Phalangiidae, Opiliones Palpatores). Senckenberg. Biol., 38(5/6) : 32 3-35 8 . Shear, W. A. 1980. Leiobunum trimaculatum Goodnight and Goodnight is a synonym of L. bimacula- tum Banks (Opiliones, Leiobunidae). J. Arachnol., 8:284-285. James C. Cokendolpher, Department of Biological Sciences, Texas Tech University, Lubbock, Texas 79409. Manuscript received March 1981, revised April 1981. The Journal of Arachnology 10:91 SURGICAL RESTRAINT APPARATUS FOR LIVING SPIDERS Experiments involving the surgical manipulation of a living spider require immobiliza- tion of the specimen. Anesthesia is a commonly used method of immobilization but the possibility exists that the chemicals used (C02 ether, chloroform, etc.) could alter the post-surgery behavior and/or physiology. Such alterations in behavior and physiology are difficult to quantify but the possibility of their existence must be assumed until data to the contrary is reported. Low temperatures have also been used to immobilize arthropods for experimental work (Berry, Miller and Harris. 1978. Ann. Entomol. Soc. Amer., 71:126-128). Dosage can be a problem with any of the above methods. Seligy (1970. Canadian J. Zool., 48:406-407) and Randall (1978. Florida Entomol., 6:192) each designed an apparatus to immobilize spiders without anesthisia for examina- tion purposes but no apparatus has been described to restrain live spiders for surgical procedures. I designed and fabricated an apparatus that holds a spider between two layers of polyurethane foam with the appendages exposed for surgery without damaging the speci- men (Fig. 1). The polyurethane allows fragile spiders (immature Latrodectus variolus Walckenaer and Peucetia viridans (Hentz) to be held, firmly immobilized, in one position without injury. A 1 1/2" wide X 3/8" disc of polyurethane is cut horizontally about three-fourths of the way through the disc. A 60 X 15 mm tissue culture dish (Falcon no. 3002) is filled to 75% of its capacity with liquid paraffin and the split polyurethane disc floated on the wax until it has hardened. Spiders as early as second instar L. variolus have been transferred from rearing con- tainers using lightweight (“mosquito”) forceps to be placed in the apparatus. Insect pins are inserted through the disc, around the spider and into the paraffin to secure the specimen in place. This apparatus, easily positioned under a dissecting microscope, has been used to immobilize spiders for amputation and ligation procedures. To facilitate surgical procedures on the prosoma and opisthosoma a modification of this device was made by cutting a hole in the top layer of the slit disc through which the operations could be performed. This devise was envisioned and fabricated at the Entomology and Nematology Depart- ment, University of Florida, Gainesville, Florida 32611. Present address: John B. Randall, 7 Twaddell Mill Road, Wilmington, Delaware 19807. Manuscript received October 1980, accepted December 1980. The Journal of Arachnology 10:92 MANTISPIDAE (INSECTA: NEUROPTERA) PARASITIC ON SPIDER EGG SACS: AN UPDATE OF A PIONEERING PAPER BY B. J. KASTON Much of the early work on the neuropteran family Mantispidae concerned the seren- dipitous rearings of adult mantispids from spider egg sacs. One of the first such mantispid- spider associations for North America was published by B. J. Kaston (1938). It is an account of the emergence of Mantispa fusicornis (sic) Banks from the egg sac of Agel- enopsis naevia (Walckenaer) (cited as Agelena naevia). The spider was collected near Albion, Michigan and transported to New Haven, Connecticut where it formed the egg sac. The mantispid was found dead in the vial several weeks later. Kaston pointed out that the larva had either crawled into the spider’s container in New Haven or had been transported from Michigan on the spider’s body. Hungerford’s (1939) observation of 10-15 first instar larvae on the pedicel of a female A rctosa littoralis (Hentz) prompted Kaston (1940) to conclude that his second suggestion was correct. The matter has re- mained closed for over 40 years. Mantispa fuscicornis is one of three sibling species occurring in North America, the other two being M. sayi Banks andM uhleri Banks (MacLeod, in Hughes-Schrader 1979). I have had M. uhleri under intensive laboratory examination for the past eight years; I have collected and reared the remaining two. All three are virtually indistinguishable on the basis of cytology (Hughes-Schrader 1979) and adult morphology. Nevertheless, they possess distinctive color patterns, are largely allopatric and, unless future investigations suggest otherwise, should be regarded as distinct species (MacLeod, in Hughes-Schrader 1979). Several factors suggested that Kaston’s specimen might actually be M. uhleri. M. fuscicornis was described from Florida (Banks 1911) and Kaston indicates a lack of records farther north than Virginia. In constrast, the holotype of M. uhleri is from Pennsylvania and there are several paratypes from Illinois and Wisconsin (Banks 1943). I have collected numerous M. uhleri adults and larvae in several Illinois locales but know of no M. fuscicornis records from this state. Similarly, of these three sibling species, Throne (1972) shows only M. uhleri occurring in Wisconsin. Thus, Michigan would seem to be a likely habitat for M. uhleri , but questionable for M. fuscicornis. Secondly, Kaston’s description of his specimen does not mention a broad, inverted Y-shaped line on the frons, the forks of the Y looping under the antennal sockets. These loops are distinctive; I have observed them only in M. fuscicornis. Since Kaston does mention the median longitudinal line on clypeus and labrum found in both species, it seems unlikely that he would have not described the loops under the antennae of M. fuscicornis. Most suggestive of a misidentification is the fact that Kaston’s paper precedes Banks’s description of M. uhleri by five years. I had occasion to discuss this with Dr. Kaston several years ago and he remembered that Banks had proclaimed the specimen as fusci- comis after a rather cursory examination under a hand lens. “Since he had described the species himself, I was not about to question him on it.” It is not possible to say whether Banks had yet formulated the idea of M. uhleri as a new species, but it would certainly not be surprising for him to have misidentified it under the circumstances. Dr. Kaston was unaware of the specimen’s fate and as of 1975 it could not be located at the Connecticut Argricultural Experiment Station (CAES). It has now happily reap- The Journal of Arachnology 10:93 peared thanks to the assistance and sharp eyes of Dr. Chris T. Maier who discovered it residing in the wrong drawer at CAES. The specimen is indeed Mantispa uhleri Banks. My laboratory and field investigations (Redborg and MacLeod, in press) corroborate Kaston’s contention that his larva boarded the spider in Michigan. It was probably on the pedicel or in one of the book lungs. M. uhleri’s host range is extremely broad; it includes nearly every family of hunting spiders (Redborg and MacLeod, in press). Although agelenids are not common hosts forM uhleri — probably because they are inaccessible in their funnel webs - I have collected one other specimen in association with an agelenid. Rather appropriately, the spider is Agelenopsis kastoni Chamberlin and Ivie. Since M. uhleri’s host range is now the most firmly documented of any mantispid species, it is also appropriate that one of the first and most widely cited papers on the life history of mantispids turns out to be the first contribution to our knowledge of this species. I thank Dr. Gilbert P. Waldbauer for critical examination of the manuscript and Dr. Ellis G. MacLeod for sharing with me the principal results of his unpublished revisionary studies of the Nearctic Mantispidae. LITERATURE CITED Banks, N. 1911. Descriptions of new species of North American Neuroptera. Trans. Amer. Entomol. Soc., 37(4): 335-360. Banks. N. 1943. New Neuroptera and Trichoptera from the United States. Psyche, 5 0( 3-4) : 7 4-8 1 . Hughes-Schrader, S. 1979. Diversity of chromosomal segregational mechanisms in mantispids (Neur- optera: Mantispidae). Chromosoma, 75:1-17. Hungerford, H. B. 1939. A note on Mantispidae. Bull. Brooklyn Entomol. Soc., 34:265. Kaston, B. J. 1938. Mantispidae parasitic on spider egg sacs. J. New York Entomol. Soc., 46:147-153. Kaston, B. J. 1940. Another Mantispa reared. Bull. Brooklyn Entomol. Soc., 35:21. Redborg, K. E. and E. G. MacLeod. The developmental ecology of Mantispa uhleri Banks (Neuroptera: Mantispidae). Illinois Biological Monographs. In press. Throne, A. 1972. The Neuroptera - suborder Planipennia of Wisconsin Part III - Mantispidae, Ascalaphidae, Myrmeleontidae and Conipoterygidae. Great Lakes Entomol., 5(4): 119-128. Kurt E. Redborg, Department of Entomology, University of Illinois Urbana, IL 61801 (present address: Department of Agronomy, University of Illinois, 1102 South Goodwin Avenue, Urbana, IL 61801) Manuscript received May 1981, revised June 1981. A NOTE ON SOME SUPPOSED TEXAN LOCALITIES FOR SOME ARANEUS SPECIES (ARANEAE, ARANEIDAE) Ever since the beginning of biogeography there has been a more or less implicit recognition of the interaction of distribution and ecology: that the range of any organism can only reflect the occurrence of suitable biotopes available to that species during its dispersal. The relationship between ecology, climate, and physiography is so pervasive that distinct biogeographic patterns shared by a host of unrelated organisms, ranging from intercontinental to extremely local, can be observed and defined. As a result, if a distri- bution map of a particular species shows one or more localities far outside the apparent The Journal of Arachnology 10:94 natural range, one is justified in regarding such records with suspicion, the more so if there is a difference in climate or physiography involved. A recent publication on a group of Nearctic araneid spiders (Levi, 1971, Bull. Mus. Comp. Zool., 141:131-179) provides an example of the situation mentioned above. Reference to the many spot maps in that paper shows clearly that most species of the Araneus diadematus group are boreal in range, extending southward (if at all) only along the major mountain systems. Only A. marmoreus and A. bicentarius occur in the Missis- sippi Valley and Gulf Coast region. Some notable exceptions to this generality may be noted, however, namely A. nord- manni (Map 3), A. cavaticus (Map 8, above), and A. illaudatus (Map 8, below), all of which are basically montane or northern in range with remotely isolated occurrences shown in Texas. Having some knowledge of A. cavaticus in the central Appalachians, I have been skeptical of its inclusion in the Texan fauna. As is well shown by the map, this spider occurs from northern Alabama to Nova Scotia. South of Pennsylvania it is characteristically found beneath high cliffs and escarp- ments in mixed mesophytic forest. Where would such an animal live in the vicinity of Houston, Texas, the locality mentioned in the text? A very similar kind of disjuction may be found in the literature on Diplopoda, and the circumstances throw some light on the Texas locality for A. cavaticus. Some decades ago, R. V. Chamberlin (1943) recorded three typically Appalachian taxa of millipeds also from Houston, all three of them highly unlikely residents of flat Texan pine woods. These records were taken from material, said to have been collected at Houston by one Russell Scott and which is still, in the Chamberlin myriapod collection (now in the U. S. National Museum). The new species of Sigmoria that he named (, S . houstoni) has subsequently been found in eastern Tennessee, and the status of the “new” spirobolid Spirobolus scotti has been discussed by Keeton (1960, Mem. Amer. Ent. Soc., 17:1-146) who found the type specimen to be only a normal individual of Narceus annularis (Rafinesque) which is not known to occur further south than Chattanooga, Tennessee. On the suspicion that something similar might appear in the case of A. cavaticus , I appealed to Dr. Norman Platnick for information on any possible Houston material in the American Museum of Natural History, where Chamberlin’s arachnid material was deposited. I was not surprised when he discovered there a pair of A. cavaticus labeled “Houston, Harris Co., Texas, Sept .-Dec. 1941, Russell Scott.” I think that the weight of evidence points to Scott’s “Houston” material being mis- labeled: either it came from some real place in Tennessee (where there is a Houston County), or was sent to Chamberlin from Houston, Texas, which he assumed to be the collection site, or most likely, there was simply some outright error in labeling on the part of Chamberlin or a preparator. Anyone familiar with Chamberlin’s modus operandi will not be surprised to learn that the myriapod material, at least, is infested with incorrect labels, and there is no reason to think he was more careful with his spiders. Lastly, a lot of material in the same collection said to have been collected at Edinburg, Texas, is patently mislabeled (including the type specimens of Costa Rican and Peruvian millipeds) and I think that the south Texas localities shown on the maps for A. illaudatus and A. nordmanni can be traced back also to Chamberlin material. In view of the ecology and distribution of these species, it seems utterly unlikely that either — especially nord- manni — exists in the lower Rio Grande valley or anywhere else in Texas. Since, as has been so rightly stated “. . .systematics and biogeography form an in- separable whole” (Wygodzinsky, 1967, Biol. l’Amer. Austr., 3:505-524), it is appropriate The Journal of Arachnology 10:95 for systematists to be on the alert for improbable disjunctions, and define or map the distribution of species with the same care that they study the characters of species. Labeling errors are lamentably frequent, but the more implausible records can be easily detected and investigated whenever biogeographic patterns are contradicted. Richard L. Hoffman, Biology Department Radford University, Radford, VA 24142. Manuscript received April 1981, revised May 1981. BOOK REVIEW Keegan, Hugh L. 1980. Scorpions of Medical Importance. Univ. Press of Mississippi, Jackson. 140 pp. ($22.50). Perusing the outside back cover the potential buyer learns that this book’s scope is to present “an account of the distribution, morphology, biology and classification of these scorpions considered to be of medical importance,” and that “A valuable feature of Scorpions of Medical Importance is the outstanding drawings that have been used to illustrate the species. It is not often that one comes across drawings so striking in their precision and attention to even the most minute details.” Between the covers the reader finds six chapters, each with its own list of references. The chapters are summarized and characterized below: Chapter 1— Scorpion Morphology and Biology. A 13 page chapter, of which five pages are plates and two are references. In general it presents a good, albeit brief review of morphology and biology. I was slightly disturbed upon reading that there are six, rather than the customary five metasomal segments plus the telson. Chapter 2— Geographic distribution of Dangerously Venomous Scorpions. A six page chapter with a one-half page of text, a four page long table, and two pages of references. The table lists, by country, the species considered of medical importance. It also includes some countries which appear to have no scorpionism problems. The table is hard to follow at times, and it’s occasionally contradicted by later statements in the text about both the medical importance ( Tityus trivittatus Kraepelin), and the distribution ( Centrur - oides sculpturatus Ewing in Mexico) of some dangerous species. Chapter 3— Clinical Aspects of Scorpion Envenomation. This eight page chapter exemplifies the problems associated with envenomation accidents in general, such as the rather variable symptomatology, and the fact that there is no general agreement as the the cause of death. It includes a one and a half page table of symptoms produced by “selected species of medical importance” (actually only four species represented), and a list of current (1978) antivenin production laboratories. Chapter 4— Scorpion Control and Prevention of Scorpion Stings. This chapter includes three and a half pages of text and a page of references touching upon preventive, mechanical, and chemical control measures. Some of the methods suggested are no longer valid: the use of chlordane (2%) against anything other than termites and when used by anyone other than a licensed pest control operator is illegal in the United States. Chapter 5— Classification of Scorpions. Two pages of text in which six families (Chaerilidae omitted) are briefly characterized morphologically. The incorrect spellings Vejovis and Vejovidae are unnecessarily perpetuated. No list of references is given with this chapter. The Journal of Arachnology 10:96 Chapter 6— Accounts of Genera and Species. A 90 page chapter, of which six are references and 53 are illustrations. Twenty-six species are each given two full pages of illustrations: one page being an entire dorsal view; the other detailing lateral and ventral views of the metasoma, the dorsoexternal aspect of the pedipalp chela, dorsal aspect of the chelicera (fingers closed), the sternopectinal area, and a ventral view of the prosoma and mesosoma. The drawings are indeed striking in their precision and are very nice pieces of artwork, but they are deemed useless for identification and/or recognition purposes. The species illustrated appear to have been chosen on the basis of their availability while the author was in Japan (where five artists prepared the drawings). Of 26 species listed in Chapter 2 as being of medical importance, only 14 are illustrated. And 12 species of little or no medical importance (perhaps other than their large size and potential for entomophobia) are included. For most arachnologists the overall usefulness of this book resides in the references provided at the end of the chapters, and the justification for such a high-priced partial bibliography is the aesthetic value of the illustrations. Oscar F. Francke, Department of Biological Sciences, Texas Tech University, Lub- bock, Texas 79409. THE AMERICAN ARACHNOLOGICAL SOCIETY President: Jonathan Reiskind (1981-1983) Department of Zoology University of Florida Gainesville, Florida 32601 Membership Secretary : Norman I. Platnick (appointed) American Museum of Natural History Central Park West at 79th Street New York, New York 10024 Secretary: William A. Shear (1980-1982) Department of Biology Hampden-Sydney College Hampden-Sydney, Virginia 23943 President-Elect: Susan E. Riechert (1981-1983) Department of Zoology University of Tennessee Knoxville, Tennessee 37916 Treasurer: Norman V. Horner (1981-1983) Department of Biology Midwestern State University Wichita Falls, Texas 76308 Directors: Herbert W. Levi (1981-1983) Michael E. Robinson (1981-1983) Vincent D. Roth (1980-1982) The American Arachnological Society was founded in August, 1972, to promote the study of the Arachnida, to achieve closer cooperation between amateur and professional arachnologists, and to publish The Journal of Arachnology. Membership in the Society is open to all persons interested in the Arachnida. Annual dues are $20.00 for regular members, $15.00 for student members. Correspondence con- cerning membership in the Society must be addressed to the Membership Secretary. Members of the Society receive a subscription to The Journal of Arachnology. In addi- tion, members receive the bi-annual newsletter of the Society, American Arachnology. American Arachnology , edited by the Secretary, contains arachnological news and comments, requests for specimens and hard-to-find literature, information about arach- nology courses and professional meetings, abstracts of the papers presented at the Society’s meetings, address changes and new listings of subscribers, and many other items intended to keep arachnologists informed about recent events and developments in arach- nology. Contributions for American Arachnology must be sent directly to the Secretary of the Society. The Eastern and Western sections of the Society hold regional meetings annually, and every three years the sections meet jointly at an International meeting. Information about meetings is published in American Arachnology , and details on attending the meetings are mailed by the host(s) of each particular meeting upon request from interested persons. The 1982 Eastern section meeting will be hosted by William A. Shear and Hampden- Sydney College, Virginia. The 1982 Western section meeting will be hosted by William F. Rapp and Doan College, Crete, Nebraska, on 28-30 July 1982. CONTENTS THE JOURNAL OF ARACHNOLOGY VOLUME 10 SPRING 1982 NUMBER 1 Feature Articles An apparatus and technique for the forcible silking of spiders, Robert W. Work and Paul D. Emerson 1 The pseudoscorpion genus Corosoma Karsch, 1879, with remarks on Dasychernes Chamberlin, 1929 (Pseudoscorpiones, Chernetidae), Volker Mahnert 11 Regression estimate of population size for the crab spider Philodromus cespitum (Araneae, Philodromidae), Joan P. Jass 15 Prey records for the Green Lynx spider, Peucetia viridans (Hentz) (Araneae, Oxyopidae), John B. Randall 19 A comparison of cursorial spider communities along a successional gradient, Thomas L. Bultman, George W. Uetz and Allen R. Brady 23 Are there any bothriurids (Arachnida, Scorpiones) in southern Africa? Oscar F. Francke 35 Epichernes aztecus, a new genus and species of pseudoscorpion from Mexico (Pseudoscorpionida, Chernetidae), William B. Muchmore and Edna Hentschel 41 Epigamic display in jumping spiders (Araneae, Salticidae) and its uses in systematics, David B. Richman 47 Predator avoidance behaviors of five species of Panamanian orb-weaving spiders (Araneae; Araneidae, Uloboridae), Debra K. Hoffmaster 69 Influence of activity patterns on social organization of Mallos gregalis (Araneae, Dictynidae), William J. Tietjen 75 Research Notes Diurnalism in Parabuthus villosus (Peters) (Scorpiones, Buthidae), Alexis Harington 85 A new Rhinochernes from Ecuador (Pseudoscorpionida, Chernetidae), William B. Muchmore 87 Comments on some Leiobunum species of the U. S. A. (Opiliones; Palpatores, Leiobunidae), James C. Cokendolpher 89 Surgical restraint apparatus for living spiders, John B. Randall 91 Mantispidae (Instecta, Neuroptera) parasitic on spider egg sacs: an update of a pioneering paper by B. J. Kaston, Kurt E. Redborg 92 A note on some supposed Texan localities for some Araneus species (Araneae, Araneidae), Richard L. Hoffman 93 Book Review Scorpions of Medical Importance by Hugh L. Keegan, Oscar F. Francke 95 Cover illustration, Alacran tartarus Francke, by Oscar F. Francke Printed by The Texas Tech University Press, Lubbock, Texas Posted at Crete, Nebraska, on April 1982 The Journal of ARACHNOLOGY OFFICIAL ORGAN OF THE AMERICAN ARACHNOLOGICAL SOCIETY VOLUME 10 SUMMER 1982 NUMBER 2 THE JOURNAL OF ARACHNOLOGY EDITOR: Oscar F. Francke, Texas Tech University. ASSOCIATE EDITOR '. B. J. Kaston, San Diego State University. EDITORIAL ASSISTANT: James C. Cokendolpher, Texas Tech University. EDITORIAL BOARD: Charles D. Dondale, Agriculture Canada. William G. Eberhard, Universidad de Costa Rica. Maria E. Galiano, Museo Argentino de Ciencias Naturales. Willis J. Gertsch, American Museum of Natural History. Neil F. Hadley, Arizona State University. B. J. Kaston, San Diego State University. Herbert W. Levi, Harvard University. Emilio A. Maury, Museo Argentino de Ciencias Naturales. William B. Muchmore, University of Rochester. Martin H. Muma, Western New Mexico University. William B. Peck, Central Missouri State University. Norman I. Platnick, American Museum of Natural History. Susan E. Riechert, University of Tennessee. Michael E. Robinson, Smithsonian Tropical Research Inst. Jerome S. Rovner, Ohio University. William A. Shear, Hampden-Sydney College. Carlos E. Valerio, Universidad de Costa Rica. Stanley C. Williams, San Francisco State University. THE JOURNAL OF ARACHNOLOGY is published in Spring, Summer, and Fall by The American Arachnological Society in cooperation with The Graduate School, Texas Tech University. Individual subscriptions, which include membership in the Society, are $20.00 for regular members, $15.00 for student members. Institutional subscriptions to The Journal are $25.00. Correspondence concerning subscription and membership should be addres- sed to the Membership Secretary (see back inside cover). Back issues of The Journal are available from Dr. William B. Peck, Department of Biology, Central Missouri State Univer- sity, Warrensburg, Missouri 64093, U.S.A., at $5.00 for each number. Remittances should be made payable to The American Arachnological Society. Change of address notices must be sent to both the Secretary and the Membership Secretary. Manuscripts for THE JOURNAL OF ARACHNOLOGY are acceptable only from cur- rent members of the Society, and there are no page charges. Manuscripts must be type- written double or triple spaced on 8.5 in. by 11 in. bond paper with ample margins, and may be written in the following languages: English, French, Portuguese, and Spanish. Contributions dealing exclusively with any of the orders of Arachnida, excluding Acari, will be considered for publication. Papers of a comparative nature dealing with chelicer- ates in general, and directly relevant to the Arachnida are also acceptable. Detailed instructions for the preparation of manuscripts appear in the Fall issue of each year, and can also be obtained from the Editor and the Associate Editor. Manuscripts that do not follow those instructions will be returned to the author(s) without the benefit of review. Manuscripts and all related correspondence must be sent to Dr. B. J. Kaston, 5484 Hewlett Drive, San Diego, California 92115, U.S.A. Benedict, E. M. and D. R. Malcolm. 1982. Pseudoscorpions of the family Chernetidae newly identified from Oregon (Pseudoscorpionida, Cheliferoidea). J. Aranchol., 10:97-109. PSEUDOSCORPIONS OF THE FAMILY CHERNETIDAE NEWLY IDENTIFIED FROM OREGON (PSEUDOSCORPIONIDA, CHELIFEROIDEA) Ellen M. Benedict and David R. Malcolm College of Arts and Sciences Pacific University Forest Grove, Oregon 97116 ABSTRACT The diagnosis of the genus 77 linichemes Hoff is revised and a new species, Illinichemes stephensi, is described from tree hollows in western Oregon; the first Oregon state records are reported for Acuminochemes crassopalpus (Hoff), Dendrochemes crassus Hoff, Dinocheirus sicarius Chamberlin, D. validus (Banks), Hesperochem.es utahensis Hoff and Clawson, and Lustrochernes grossus (Banks), and Lamprochernes sp. These are the first published records from Oregon for the chernetid family. INTRODUCTION Chernetid pseudoscorpions have not been reported previously from Oregon even though approximately 80 species of the family are known from other parts of the United States. A recent study of Oregon pseudoscorpions (Benedict 1978), based on newly collected specimens and on specimens accumulated in various older collections, indicates that chernetids are fairly common in Oregon habitats. Some of the specimens were readily identifiable as: Acuminochemes crassopalpus (Hoff), Dendrochemes crassus Hoff, Dinocheirus sicarius Chamberlin, D. validus (Banks), Hesperochernes utahensis Hoff and Clawson, and Lustrochernes grossus (Banks). Follow- ing examination of the type series of Illinichemes distinctus Hoff, the type species of the genus, Oregon specimens of Illinichemes were determined to represent a new species which is described in this paper. However, other specimens were more difficult to identify because many of the genera and species of this family are poorly defined. For example, certain specimens, clearly assignable to the genus Dinocheirus Chamberlin, presented complex problems; Muchmore (1974) has begun a comparative study of this genus which, when complete, should permit the identification of the remaining Oregon specimens of this taxon. Further, the single known specimen of the genus Lamprochernes Tomosavary collected in Oregon is not identifiable to species at this time. Recent contributions of such workers as Muchmore (1974, 1975), Legg (1974a, 1974b, 1975), Weygoldt (1966, 1970), and others, may ultimately provide the basis for a major revision of the family. The need for a broad critical study with modern descriptions of all included genera and species is apparent. 98 THE JOURNAL OF ARACHNOLOGY A number of works, in addition to those cited above, contain major contributions to the present, though incomplete, knowledge of the chernetids. Chamberlin (1931) pro- vided numerous illustrations in his comparative morphological monograph of the order; Beier (1932) a brief description with general distributions of the then recognized 190 species of the world together with a descriptive key to many of the species; Beier (1963) a descriptive key to 45 European species; and Hoff (1958) a list of 71 chernetid species from the United States and Canada together with a useful key to genera. The seven genera identified from Oregon may be distinguished by the following key: 1 . Tibia of leg IV with tactile seta; pleural membrane smoothly striate 2 Tibia of leg IV lacking a tactile seta; pleural membrane not smoothly striate . . 3 2. Tibia of leg IV with two tactile setae, one distal and one near middle of segment; tactile seta IT of fixed chelal finger at least as close to finger tip as distance between 1ST and ISB Lustrochernes grossus Tibia of leg IV with a single tactile seta distal in position; tactile seta IT distinctly farther from finger tip than distance between 1ST and ISB . . Lamprochernes sp. 3. Tarsus of leg IV with a tactile seta 4 Tarsus of leg IV without a tactile seta 7 4. Movable chelal finger with tactile seta ST closer to SB than to T Dendrochernes crassus Movable chelal finger with tactile seta ST midway between T and SB or closer to T than to SB 5 5. Cheliceral hand with seta sb and b acuminate . . . Acuminochernes crassopalpus Cheliceral hand with seta sb denticulate and seta b acuminate . . (Dinocheirus) 6 6. Palpal femur of male with large protuberance on subdorsal (inner) surface; tactile seta of tarsus IV greater than 70% of the total tarsal length from proximal margin . D. sicarius Palpal femur of male without a large protuberance on subdorsal surface; tactile seta of tarsus IV less than 65% of the total tarsal length from proximal margin . . . . D. validus 7. Setae of palps and tergites bilaterally feathered and leaflike; several extra-long clavate setae near center of outer margin of fixed finger . . Illinichernes stephensi Setae of palps not bilaterally feathered; without extra-long clavate setae near center of outer margin of fixed finger Hesperochernes Acuminochernes Hoff Hoff (1949) erected this genus and designated his species Hesperochernes crassopalpus the type species, which he had initially described in great detail from specimens collected in Arkansas (Hoff 1945). In 1949, he provided additional measurements of the palp and illustrations of the palp and chela from specimens collected in Illinois and Kansas. Hoff (1961) distinguished A. crassopalpus from a newly described species, A. tactitus Hoff, collected in Colorado. Nelson (1975) gave further measurements for ,4. crassopalpus from Michigan specimens. Recently (1981) Muchmore has concluded that Phoberocheirus from the southern United States is a junior synonym of Acuminochernes. BENEDICT AND MALCOLM-CHERNETID SPECIES FROM OREGON 99 Acuminochernes crassopalpus (Hoff) The single known specimen of the genus Acuminochernes from Oregon is identifiable as A. crassopalpus (Hoff 1949, 1958). Although no specimens of this species have been reported from localities between Oregon and the Great Plains states, it is probable that this disjunct distribution reflects a lack of collections rather than the true situation. New record .-Oregon: Columbia Co., Sauvies Island, 5 mi N, 2 mi E of Burlington (near sea level), debris of Spermophilus beecheyi douglasii (Richardson) in hollow of mature Quercus garryana Dougl., 7 October 1972 (E. M. Benedict), 1 male (EMB). Dendrochernes Beier The genus Dendrochernes Beier was erected in 1932, at which time Chemes cyrnus L. Koch from Europe was designated the type species. Hoff (1949, 1956) has characterized the genus in English and included it in his key (1958). The genus is known from the United States only from three species and relatively few localities. Dendrochernes morosus (Banks) is reported from only a few Michigan specimens (Banks 1895, Manley 1969, Nelson 1975). Dendrochernes instabilis (Chamberlin) from Montana is as poorly known (Chamberlin 1934). Dendrochernes crassus Hoff was described in detail from three specimens collected at three localities in New Mexico (Hoff 1956) and further characterized from an additional specimen from Colorado (Hoff 1961). Dendrochernes crassus Hoff Twenty-two specimens of this species have been identified from widely scattered localities in Oregon and are listed below. It appears that D. crassus typically inhabits the bark of coniferous logs or snags, usually of Pinus ponderosa Dougl. ex Loud. (Benedict 1978); unfortunately habitat data were not recorded for all collections. New records .-Oregon; Baker Co., Dooley Mt., 14 July 1958 (J. Baker), 2 females (WBM); Benton Co., 2 mi N of Corvallis, under bark of burned Pseudotsuga menziesii (Mirb.) Franco, 1938 or 1939 (J. D. Vestres), 1 male (JCC); Crook Co., near Prineville, Ochoco National Forest, under bark of Pinus ponderosa in association with Dendroctonus brevimanus Lee., no date (W. J. Buckhorn), 2 males, 2 females (JCC), Prineville, 27 April 1933 (W. J. Buckhorn), 1 male, 3 females (WBM), near Prineville, Ochoco Ranger Station, 12 March 1939 (collector unknown), 2 males, 1 female (JC C); Douglas Co., 6 mi W of Glide, bark of log ofP. ponderosa, 1 April 1972 (E. M. Benedict), 1 female (EMB);Z,m« Co., Cascadia, 9 May 1939 (S. Jewett), 1 female (JCC); Wasco Co., Schoolie Ranger Station, Warm Springs Indian Reservation, bark of P. ponderosa, 4 July 1938 (J. C. Chamberlin, R. L. Prentiss, C. Prentiss), 2 males, 2 females (JCC); Washington Co., Timber, bark of dead Pseudotsuga menziesii, no date (R. L. Furniss), 2 females (JCC). Dinocheirus Chamberlin The genus Dinocheirus , which Chamberlin (1929) erected for his new species Dinocheirus tenoch Chamberlin from Mexico, was recently re-diagnosed by Muchmore (1974) and clearly distinguished from the genera Chemes Menge and Hesperochernes Chamberlin. Of the 19 species from the United States then apparently assigned to Dinocheirus (see Hoff 1958, Nelson and Manley 1972), Muchmore stated that he was confident, on the basis of either an adequate original description or the re-examination of type specimens, that eight of these species clearly belonged to the genus Dinocheirus and 100 THE JOURNAL OF ARACHNOLOGY two belonged to another genus. Further, he felt “uncertainty” about eight species not re-examined. Unfortunately, Muchmore made no mention of the fact that one of the species of Dinocheirus listed by Hoff in 1958, D. stercoreus Turk, had been transferred by Hoff in 1957 to the genus Tejachernes Hoff as its type species. The status of this species will remain confused until the validity of the genus Tejachernes is determined. Although Muchmore’s study is most helpful, a comparative redescription of all of these species based upon re-examination of the type specimens and large series of additional specimens is needed before the individual species can clearly be distinguished. In the meantime, it is possible to assign some of the Oregon specimens of the genus Dinocheirus to two of these species: Dinocheirus sicarius Chamberlin and D. validus (Banks). Dinocheirus sicarius Chamberlin Chamberlin (1952)' described this species in great detail, complete with excellent illustrations from specimens collected from habitats associated with domestic animal shelters in California. Although Gering (1956) listed the species from the Great Salt Lake Desert of Utah, he provided neither specimen records nor habitat data. The first specimen records for Oregon are now reported from 12 localities in nine western Oregon counties where D. sicarius typically occurs in the litter-dung layer of cow, horse, sheep, pig and chicken sheds and barns. Oregon specimens appear as variable in size as those described by Chamberlin from California. New records.— Oregon; Benton Co ., 1.3 mi NW Summit (150 m), litter-dung of cattle-sheep barn, 20 December 197 1 (E. M. Benedict), 5 males, 3 females, 10 nymphs (EMB); Columbia Co., 3 mi SE of Clatskanie (100 m), litter-dung of chicken house, 8 January 1972 (E. M. Benedict), 6 males, 3 females, 1 nymph (EMB); Coos Co., 5 mi N, 1 mi E of Langlois (60 m), litter-dung of cattle barn, 27 July 1973 (E. M. Benedict), 4 females with eggs, 10 nymphs (EMB); Jackson Co., 3 mi E of Ashland (610 m), litter-dung of cattle barn, 27 December 1971 (E. M. Benedict), 1 male (EMB); Lane Co., 5 mi N of Elmira (130 m), litter-dung of chicken house, 4 December 1971 (E. M. Benedict), 1 female (EMB), 5 mi N of Elmira (130 m), litter-dung of cattle-sheep-pig barn, 4 December 1971 (E. M. Benedict), 2 females (EMB); Lincoln Co., 1 mi NW of Elk City, (60 m), litter-dung of cattle barn, 20 December 1971 (E. M. Benedict), 8 males, 3 females (EMB); Tillamook Co., 5 mi SE of Blaine (150 m), litter-dung of cattle barn, 15 March 1972 (E. M. Benedict), 5 males, 5 females, 10 nymphs (EMB); Yamhill Co., 0.5 mi S of Yamhill (60 m), litter-dung of horse barn, 1 January 1972 (E. M. Benedict), 2 males, 2 females, 3 nymphs (EMB), 2 mi S of Carlton (60 m), litter-dung of cattle barn, 1 January 1972 (E. M. Benedict), 3 males, 3 females, 3 nymphs (EMB), 6 mi W of Carlton (200 m), litter-dung of cattle barn, 12 May 1973 (E. M. Benedict), 3 males, 4 females with eggs, 10 nymphs (EMB); Washington Co., 6 mi N of Beaverton, rotted wood in cattle barn, 30 September 1964 (D. R. Malcolm), 2 males, 1 female (DRM). Dinocheirus validus (Banks) Originally described in 1895 by Banks as Chelanops validus, this species was reassigned to the genus Dinocheirus by Beier (1932). Hoff (1947, 1956, 1961) subsequently pro- vided an extensive redescription and measurements. More recently, Muchmore confirmed the generic assignment of this species to Dinocheirus. This predominantly bark-inhabiting species has been reported from widely scattered localities in California, Colorado, New Mexico, and Utah (Banks 1895, Hoff 1956, 1961, Knowlton 1972). The first Oregon specimen records are reported below. BENEDICT AND MALCOLM-CHERNETID SPECIES FROM OREGON 101 New records .-Oregon; Benton Co., Corvallis, no date (George Ferguson), 1 male (JCC), 10 mi N of Corvallis, hollow of mature Quercus garryana, 10 January 1958 (J. Lattin), 1 male, 1 female, 1 tritonymph (EMB); Douglas Co., 6.5 mi NE of Idleyld Park (305 m), hollow of mature Abies concolor (Gord. & Glend.) Lindl. ex Hieldebr., 16 August 1973 (E. M. Benedict), 1 tritonymph (EMB), 0.5 mi S of Elkton (60 m), hollow of mature Abies grandis (Dougl.) Lindl., 15 September 1973 (E. M. Benedict), 1 male, 1 female (EMB); Lincoln Co., 0.5 mi NW of Elk City (60 m), 20 December 1971 (E. M. Benedict), 1 male (EMB); Washington Co., Forest Grove, on fly in rotary flight trap (upper net 5 ft.), 1 female (JCC); Yamhill Co., 6 mi SW of Carlton (150 m), hollow of mature Quercus garryana , 1 January 1972 (E. M. Benedict), 1 female (EMB). Dinocheirus spp. Additional specimens of this genus collected in Oregon are currently unassignable to species due to inadequate descriptions of certain species said to belong to this taxon. Hesperochernes Chamberlin Chamberlin (1924) established the genus Hesperochernes , designating his new species Hesperochernes laurae Chamberlin from California as the type species. Through the years, 15 species from the United States were added to the genus, mostly through the work of Hoff and his co-workers (Hoff 1947, 1958, Hoff and Clawson 1952). Muchmore (1974) redefined the characteristics of Hesperochernes based upon re-examination of the type series of H. laurae. In that paper he confirmed six of the species as belonging to Hesperochernes and four as belonging to Chernes and expressed uncertainty about the generic assignment of six of the species. He also transferred two additional species to Hesperochernes. Comparative redescription of all of these species based on examination of type series and large series of conspecific specimens will facilitate the identification of pseudoscorpions in this group. Hesperochernes utahensis Hoff and Clawson Hesperochernes utahensis was described by Hoff and Clawson (1952) and its generic assignment confirmed by Muchmore (1974). The species has been reported previously as an inhabitant in litter of various types in semiarid areas of the Rocky Mountain states. The Oregon specimens, listed below, came from similar habitats; they were recovered from the semiarid western juniper zone of southeastern Oregon. New iQQQi&s.— Oregon; Harney Co., Diamond Craters, 13 mi S,6 mi W of Princeton (1280 m), litter of Juniperus occidentalis Hook., 14 July 1972 (E. M. Benedict), 6 males, 1 female, 5 nymphs (EMB), 27 mi N, 13 mi W of Frenchglen (1280) m), litter of Artemesia tridentata Nutt, and Grayia spinosa (Hook.) Moq., 8 September 1976 (E. H. Gruber and E. M. Benedict), 2 males (EMB), Alvord Desert Sand Dunes, T36S, R35E, sec. 8 NW (1325 m), litter of Sarcobatus vermiculatus (Hook.) Torr., 14 October 1979 (E. H. Gruber), 1 male (EMB). Illinichernes Hoff Illinichernes Hoff 1949:481, 1958:25; Lawson 1968: 192; Nelson 1975:290. Revised diagnosis.— Cheliceral flagellum with four setae: two long and two short; hand with five or six setae (accessory seta absent or present); sb and a denticulate, b acuminate 102 THE JOURNAL OF ARACHNOLOGY or denticulate. Palps stout, exhibiting little sexual dimorphism. Setae, especially on palps and dorsum of body, bilaterally feathered, leaflike and stout; setae of sternal scuta chiefly clavate; proximal 2/3 of fixed finger bearing several prominent enlarged, long, leaflike setae. Tactile seta ST of movable chelal finger closer to T than to SB; 1ST considerably distad of EST on fixed chelal finger; both IB and ISB distad of ESB. Tarsus of leg IV without tactile seta. Paired spermathecae long slender tubules terminating in enlarged bulbous sacs. Type species.— Illinichernes distinctus Hoff. Remarks.— The genus Illinichernes was erected by Hoff (1949) for the single new species, I. distinctus , from Illinois. He emphasized the acuminate nature of seta b on the cheliceral hand and the distinctive leaflike setae on the body and palps. Subsequent workers (Hoff and Bolsterli 1956, Lawson 1968, Nelson 1975) added descriptive data and records from Indiana, Maryland and Michigan. They also reiterated the nature of b and the leaflike setae. During the past decade, pseudoscorpions were collected from tree hollows in Oregon which resembled Illinichernes in their leaflike setae, but differed in that seta b is denticulate. Further, these specimens possess a denticulate accessory seta on the cheliceral hand, a characteristic not mentioned by previous workers. In an effort to assign the Oregon specimens to a genus, types of I. distinictus were examined. The types do, indeed, bear an acuminate seta b and lack an accessory seta just as Hoff described them. Interestingly, Hoff and Clawson (1952) had questioned the value of b as a diagnostic character for separating chernetid genera; the present authors agree. However, the presence of enlarged leaflike setae on the palps and body appear adequate to discriminate the genus Illinichernes from the very closely related genus Hesperochernes without reference to the denticulate or acuminate nature of seta b. Although Hoffs diagnosis of Illinichernes was generally satisfactory, it has been necessary to make a few changes to reflect the variation of the new species from Oregon. The major changes are that: (1) an accessory seta may be present on the cheliceral hand, and (2) seta b may be either denticulate or acuminate. Spermathecae are as illustrated. Illinichernes stephensi, new species Figures 1-8 Etymology.— This species is named for Charles L. Stephens, the father of the senior author, who has been instrumental in the collecting of numerous Oregon specimens of pseudoscorpions. Type records.-Oregon; Douglas Co., 2 mi E of Canyonville (380 m), hollow of mature Abies concolor , 6 November 1971 (E. M. Benedict and C. L. Stephens), 2 males (holotype AMNH, 1 paratype EMB), 4 females (allotype AMNH, 3 paratypes EMB), 2 nymphs (paratypes EMB), 13 September 1973, 3 males, 1 nymph (paratypes EMB). Distribution.— Reported only from western Oregon. Diagnosis.— Based on adults. Carapace length of male 0.82-0.92 mm, of female 0.88-1.05 mm; palpal femur length of male 0.72-0.81 mm, of female 0.69-0.89 mm; cheliceral hand with a denticulate accessory seta (total of 6) and several elongate setae near the center of the outer margin on the proximal 1/3 of the fixed finger of the chela. Description.— Measurements in Table 1, morphometric ratios in Table 2. Moderately large, blind epigean species; derm mostly granular throughout; setae stout, pinnately feathered or leaflike to clavate, except as noted. BENEDICT AND MALCOLM-CHERNETID SPECIES FROM OREGON 103 MALE. Carapace (Fig. 1): subtriangular, longer than posterior breadth, with two distinct transverse furrows, both broad, well defined, and laterally procurved; eyes or ocular spots absent; approximately 110 stout, leaflike setae, holotype with 7 setae on anterior margin and 17 setae on posterior disc. Coxal area: subapical seta of maxilla much longer than apical seta. Abdomen (Fig. 1): ovate; tergites 1-10 and sternites 4-10 divided; pleural membrane hispidously wrinkled, interscutal and intersegmental membranes more or less hispid; setae pinnately feathered, leaflike (Fig. 2) to denticulo-clavate in some degree, except for the acuminate setae on the genital opercula, spiracular plates and lateralmost seta on sternite 5; tergal setae staggered in 1-2 irregular marginal rows and a single lateral discal seta per scutum, holotypic tergal chatotaxy 17: 19:21 :23:24:25:23:24:23:20:12:mm; sternal setae in 1-2 irregular rows, generally with a single marginal row plus a single medial discal seta per scutum, holotypic sternal chaetotaxy 41 ±:(0-0):(l )2-3/20(2):(2)l 3(3):21 :22: 22?:22?:20: 17: 10:mm. Chelicera (Fig. 3): slightly less than 1/3 as long as carapace; hand with 6 setae (sb,b, and accessory setae all denticulate); flagellum with 2 long and 2 short blades, the distal- most blade with approximately 7 denticles along the margin, other blades acuminate; galea with 5-6 small rami; serrula exterior with 17 ligulate teeth; serrula interior with dentate apical process, 3 subapical lobes and an undivided membrane proximally; fixed Fig. l.-Minichemes stephensi, new species: dorsal view. 104 THE JOURNAL OF ARACHNOLOGY finger with 3 very tiny denticles on margin of apical tooth and 6 tiny subapical denticles; movable finger with long, slender, apical tooth and 3 subapical lobes or teeth. Palp (Fig. 1): very stout; chela with dentition, chaetotaxy and venom apparatus as illustrated (Fig. 4); marginal teeth contiguous and numbering approximately 35-36 on each finger; movable finger with 7 external and 2 internal accessory teeth, nodus ramosus of venom duct opposite teeth 19-21 (slightly nearer ST than T); fixed finger with 6 external and 3 internal accessory teeth. Legs (Fig. 1): derm except for tarsi scalelike to granulate; setae except for a few on distal portion of tarsi multidenticulate and clavate to subclavate; tarsus IV lacking a tactile seta. FEMALE. Similar to male except as noted. Slightly more robust. Abdomen with acuminate setae on genital opercula, spiracular plates, sternite 4 and medially on sternite 5; allotypic tergal chaetotaxy 18:18:19:24:23:20:22:21:22:19:14:mm; sternal chaetotaxy 1 7:(1)39(1):(2)1 2(2): 18:21 : 23:23 : 2 1 : 20: 10:mm; spermathecae as illustrated in Fig. 5. Chaetotaxy of chela as in male; fixed finger with 30-35 marginal teeth, 6-8 external teeth and 2-3 internal teeth; movable finger with 34-40 marginal teeth, 3-7 external teeth and 1-3 internal teeth. TRITONYMPH. Similar to adult except as noted. Slightly smaller and paler; derm slightly less sclerotized, especially the pedicels of the palpal podomers. Chaetotaxy (EB- 1408.01004) of carapace 7-12 (74-), of tergites 13:13:14:14:13:14:14: 14: 15:14:6:mm; of sternites 8:(l)9(l):(2)9(l):12:13:15:14:15:12:6:mm. Chaetotaxy of chela (Fig. 6) typical of species and nymphal stage; fixed finger with 28-29 marginal teeth and 3-4 external accessory teeth, internal accessory teeth not observed; movable finger with 30-33 marginal teeth and 2-3 external accessory teeth, internal accessory teeth not observed. Fig. l.-Illinichernes stephensi, new species: 2, abdominal seta; 3, chelicera; 4, lateral view of chela of male; 5, spermatheca of female (other one obscured); 6, lateral view of chela of tritonymph; 7, lateral view of chela of deutonymph; 8, lateral view of chela of protonymph. BENEDICT AND MALCOLM-CHERNETID SPECIES FROM OREGON 105 Table 1. -Measurements (in mm) of Ulinichernes stephensi, new species from western Oregon (abbreviations: R^breadth, L=length, *=length exclusive of pedicel). Male (n=14) Female (n=14) Tritonymph (n=3) Deutonymph (n=3) Protonymph (n=l) Body L 2.16-2.59 2.38-3.32 1.87-1.97 1.57-1.60 1.39 Abdominal B 1.06-1.24 1.15-1.70 0.80-0.86 0.54-0.59 0.66 Carapace L 0.82-0.92 0.88-1.05 0.69-0.71 0.54-0.59 0.45 Posterior B 0.72-0.79 0.70-0.98 0.62-0.66 0.47-0.52 0.46 Chelicera L 0.24-0.26 0.24-0.28 0.19-0.21 0.17-0.18 9 Chelicera B 0.12-0.14 0.13-0.15 0.10-0.12 0.09-0.10 ? Pedipalps Trochanter L 0.43-0.50 0.45-0.52 0.34-0.35 0.27-0.28 0.17 Trochanter B 0.25-0.30 0.25-0.32 0.19-0.20 0.15-0.16 9 Femur L 0.72-0.81 0.69-0.89 0.52-0.56 0.40-0.41 0.28 Femur B 0.29-0.35 0.27-0.33 0.20-0.21 0.15-0.16 9 Tibia L 0.70-0.79 0.66-0.89 0.47-0.52 0.38-0.39 0.22 Tibia B 0.32-0.36 0.28-0.40 0.23-0.24 0.18-0.19 9 ♦Chela L 1.05-1.11 1.07-1.27 0.82-0.84 0.66-0.67 0.47 Pedicel L 0.10-0.12 0.09-0.13 0.06-0.07 0.05-0.07 0.04 Chela B 0.35-0.45 0.39-0.47 0.29-0.30 0.23-0.24 0.17 Chela D 0.36-0.41 0.35-0.47 0.29-0.30 0.23-0.24 0.17 Movable finger ] L 0.54-0.63 0.56-0.70 0.43-0.45 0.36-0.37 0.21 ♦Hand L 0.46-0.56 0.52-0.61 0.40-0.42 0.28-0.29 0.24 Leg I Entire femur L 0.46-0.53 0.46-0.55 0.37-0.39 0.27-0.29 9 Entire femur D 0.14-0.18 0.14-0.17 0.11-0.12 0.09-0.10 9 Tibia L 0.32-0.39 0.33-0.41 0.26-0.27 0.19-0.20 9 Tibia D 0.09-0.11 0.09-0.10 0.08 0.07 9 Tarsus L 0.35-0.43 0.39-0.43 0.28-0.31 0.24-0.27 9 Tarsus D 0.06-0.07 0.07-0.09 0.06-0.07 ,0.05-0.06 9 Leg IV Entire femur L 0.65-0.72 0.66-0.76 0.47-0.48 0.38-0.39 0.30 Entire femur D 0.14-0.20 0.16-0.19 0.14-0.15 0.11 0.09 Tibia L 0.51-0.63 0.50-0.63 0.38-0.39 0.28-0.29 0.20 Tibia D 0.10-0.14 0.09-0.13 0.10-0.12 0.08-0.09 0.07 Tarsus L 0.43-0.48 0.42-0.50 0.30-0.31 0.27-0.28 0.26 Tarsus D 0.08-0.09 0.08-0.09 0.07-0.09 0.07 0.06 DEUTONYMPH. Similar to adult except as noted. Smaller and paler; derm much less sclerotized, especially the pedicels of the palpal podomeres. Chaetotaxy (EB-1 408.0 1003) of carapace 6-8(54), of tergites 10:10:10:10:10:1 1:9: 10:10: 10:6:mm, of sternites 0:(l)4(l):(2)6(2):10:10:10:10:10:9:6:mm. Cheliceral hand with 5 setae of which both sh and Z?|are;denticulate (accessory seta lacking); galea with 4 very weak rami. Chaetotaxy of chela (Fig. 7) typical of species and nymphal stage; fixed finger with 26 marginal teeth, 1 external accessory tooth and 1 internal accessory tooth; movable finger with 26 mar- ginal teeth and 1 external tooth, internal accessory teeth not observed. PROTONYMPH. Similar to adult except as noted. Much smaller and paler; derm much less sclerotized. Chaetotaxy (EB-1408.01001) of carapace 6-6(28), of tergites 6:6:6:6:6:6:6:6:6:6:4:mm, of sternites 0:(0)2(0):(l)4(l):6:6:6:6:6:6:4:mm. Cheliceral hand with 4 setae of which b is acuminate ( sb and accessory seta lacking); galea with 3 106 THE JOURNAL OF ARACHNOLOGY Table 2. -Morphometric ratios of adult Ulinichemes stephensi, new species from western Oregon (Abbreviations: B=breadth, D=depth, L=length, *=length exclusive of pedicel). Male (n-14) Female (n=14) Pedipalp Femur L/B 2. 3-2.5 2.5-2. 9 Tibia L/B 2.0-2. 3 2. 3-2.6 Chela L/D 2.5-2.1 2. 6-2.9 * Chela L/D 2. 6-2. 9 2. 8-3.0 Movable finger L/Hand L 1.0-1. 3 1 .0-1.2 * Hand L/B 1. 1-1.3 1. 3-1.5 Leg I Entire femur L/D 2. 9-3. 4 2. 7-3. 2 Tibia L/D 3. 4-4.0 3. 6-4.1 Tarsus L/D 5. 3-5. 9 5. 2-6.0 Leg IV Entire femur L/D 3.8-4. 3 3.8-4. 1 Tibia L/D 4. 3-4.9 4.5-5. 1 Tarsus L/D 5. 1-5.8 5. 1-5 .7 very weak rami. Chaetotaxy of chela (Fig. 8) typical of species and nymphal stage; fixed finger with 19 marginal teeth and no accessory teeth; movable finger with 21 marginal teeth and 1 external accessory tooth. Remarks.— This new species is morphologically very similar in most respects to I. distinctus but differs in size and cheliceral chaetotaxy. Most specimens of I. stephensi possess an accessory seta on each cheliceral hand in distinction to the regular absence of this seta on I. distinctus. However, occasionally one hand of the chelicera of I. stephensi may lack this characteristic accessory seta while the other chelicera bears the typical number of 6. Specimens of the described species of Ulinichemes can be distinguished by the combination of characters in the following couplet: Hand of chelicera with 5 setae; carapace length of male 0.65-0.72 mm, of female 0.72-0.79 mm; palpal femur length of male 0.59-0.69 mm, of female 0.60-0.78 mm; chela length of male 0.82-0.96 mm, of female 0.82-0.98 mm; from the eastern United States I. distinctus Hand of chelicera with 6 setae; carapace length of male 0.82-0.92 mm, of female 0.88-1.05 mm; palpal femur length of male 0.72-0.81 mm, of female 0.69-0.89 mm; chela length of male 1.05-1.1 1 mm, of female 1.07-1.27 mm; from Oregon . . . . I. stephensi Habitat.- Found primarily in tree hollows of various species and occasionally in associated leaf litter. Other specimens examined. —Oregon; Curry Co., 14 mi E of Gold Beach (185 m), hollow of mature Lithocarpus densiflora (H. & A.) Rehd., 10 March 1972 (E. M. Benedict), 2 males, 6 nymphs (EMB), 7 mi N, 6 mi E of Brookings (60 m), hollow of mature L. densiflora, 24 August 1973 (E. M. Benedict), 1 male (EMB), 7 mi N, 7 mi E of Brookings (90 m), hollow of mature L. densiflora, 24 August 1973 (E. M. Benedict), 3 males, 1 female, 5 nymphs (EMB); Coos Co., 19 mi N, 1 mi E of Agness (90 m), hollow of mature Acer macrophyllum Pursh, 19 February 1972 (E. M. Benedict), 13 males, 12 females, 10 nymphs (EMB); 19 mi N, 1 mi E of Agness (90 m), hollow of mature Umbellularia californica (H. and A.) Nutt., 19 February 1972 (E. M. Benedict), 2 males, 3 females, 16 nymphs BENEDICT AND MALCOLM-CHERNETID SPECIES FROM OREGON 107 (EMB), 19 mi N, 1 mi E of Agness (90 m), hollow of mature U. californica, 19 February 1972 (E. M. Benedict), 1 male, 7 females, 47 nymphs (EMB); Dowlas Co., 0.7 miW of Scottsburg (90 m), hollow of mature U. californica, 11 October 1971 (E. M. Benedict), 1 female, 1 nymph (EMB), 2 mi N of Melrose (121 m), hollow of Arbutus menziesii snag, 7 Febraury 1972 (E. M. Benedict), 3 females, 17 nymphs (EMB), 2 mi N of Melrose (121 m), hollow of mature Acer macrophyllum, 7 Feburary 1972 (E. M. Benedict), 11 males, 5 females, 21 nymphs (EMB), 3 mi N, 3 mi W of Umpqua (100 m), hollow of mature Abies grandis, 7 February 1972 (E. M. Benedict), 4 males, 18 females, 81 nymphs (EMB), 6.5 mi NE of Idleyld Park (305 m), hollow of mature Acer macrophyllum, 1 April 1972 (E. M. Benedict), 12 males, 10 females, 46 nymphs (EMB), 6.5 mi NE of Idleyld Park (305 m), hollow of mature Abies concolor, 16 August 1973 (E. M. Benedict), 1 female (EMB), 6.5 mi NE of Idleyld Park (305 m), hollow of matur e Acer macrophyllum, 16 August 1973 (E. M. Benedict), 5 males, 7 females, 19 nymphs (EMB), 1/2 mi S of Elkton (60 m), hollow of mature A bies grandis, 15 September 1973 (E. M. Benedict), 1 male, 19 nymphs (EMB); Jackson Co., 8 mi S, 13 mi E of Ashland (1650 m), leaf litter of Quercus garryana, 15 October 1972 (E. M. Benedict), 1 female (EMB)’, Lane Co., 14 mi S of Oakridge (550 m), hollow of mature Acer macrophyllum, 4 March 1972 (E. M. Benedict), 2 males, 2 females, 9 nymphs (EMB), 24 mi E of Florence (90 m), hollow of mature A. macrophyllum, 6 May 1972 (E. M. Benedict), 1 female, 11 nymphs (EMB), 6 mi W, 1 mi N of Lorane (245 m), hollow of mature A. macrophyllum, 21 July 1973 (E. M. Benedict), 2 females (1 with ? number of eggs) (EMB), 13 mi W, 2 mi N of Lorane, hollow of mature A. macrophyllum, 21 July 1973 (E. M. Benedict), 3 males, 3 females, 2 nymphs (EMB), 3 mi N, 8 mi E of Lowell (335 m), hollow of mature Alnus rubra Bong., 30 August 1973 (E. M. Benedict), 1 male, 1 female, 15 nymphs (EMB); Lincoln Co., 25 mi E. of Waldport (120 m), hollow of mature Acer macrophyllum, 5 May 1973 (E. M. Benedict), 3 males, 7 females (EMB)', Polk Co., 2.2 mi W of Falls City (245 m), hollow of mature A. macrophyllum, 11 June 1973 (E. M. Benedict), 3 males, 4 females (with approximately 20 larvae, 1 with 13 eggs), 2 nymphs (EMB); Washington Co., 3 mi SW of Tualatin, leaf litter of Arbutus menziesii, 1 January 1972 (E. M. Benedict), 1 female (EMB); Yamhill Co., 6 mi SW of Carlton (150 m), hollow of mature Quercus garryana 1 January 1972 (E. M. Benedict), 1 female (EMB), 2 mi S of Carlton (60 m), hollow of mature Q. garryana, 1 January 1972 (E. M. Benedict), 3 males, 3 females, 5 nymphs (EMB). Lamprochernes Tomosvary At present, the genus Lamprochernes is in need of revision to clearly define its limits. Recently, Muchmore (1975:19) recognized that Pycnochernes Beier is a synonym of Lamprochernes “unless further work can demonstrate that Che lifer (Chernes) god freyi Kew is not congeneric with Chelifer nodosus Schrank, the type species of Lamprochernes .” Shortly thereafter, Muchmore (1976:152) transferred four of the five American species which had earlier been assigned to Lamprochernes (Beier 1932, Hoff 1944, 1949, Hoff and Bolsteri 1956, Muchmore 1971) to the new genus Americhemes. He also provided a key by which “ Americhemes could be distinguished from Lampro- chemes and allied American genera.” By Muchmore’s criteria, it appears that Lampro- chernes is presently represented in the United States by three species: Lamprochernes minor Hoff from the midwestern states, Pycnochernes linsdalei Chamberlin from Cali- fornia and Pynochernes foxi Chamberlin from Idaho. The latter two species had been described in 1952. The single known Oregon specimen of Lamprochernes is not identi- fiable to species at this time. Material examined.-Oegcw; Douglas Co., Dixonville, phoretic on housefly, 29 September 1973 (collector unknown), 1 female (WBM). Lustrochernes Beier At present, two species of Lustrochernes are recognized from the United States: L. pennsylvanicus (Ellingsen) in the southeast and L. grossus (Banks) in the west. 108 THE JOURNAL OF ARACHNOLOGY Lustrochemes grossus (Banks) Banks (1893) described this species as Chelanops grossus from specimens collected in Colorado. Hoff (1947) redescribed the species from the types and transferred it to the genus Lamprochernes . Subsequently, upon examination of numerous specimens from New Mexico, he reassigned the species to the genus Lustrochemes (Hoff 1956). It has also been reported from Arizona. New records.-Oe^ow; no data, under elytra of cerambycid beetle (W. J. Chamberlin), 2 females (JCC); Benton Co., Corvallis, stump of Pseudotsuga menziesii, 26 April 1936 (J. C. Chamberlin), 1 male, 1 female (JCC), Corvallis, from cerambycid beetles, Ergates spiculatus spiculatus Lee., 4 August 1977 (E. Vogel), 5 females (EMB); Douglas Co., 2.5 mi E of Coos Co. line on Roseburg-Coquille Rd., rotting log of P. menziesii, 28 April 1936 (J. C. Chamberlin), 1 male (JCC); Crook Co., Prineville, 12 March 1939 (collector unknown), 1 female (JCC); Josephine Co., 15.7 mi N of Medford on Crater Lake Rd, bark oiPinus stump, 1937 (J. C. Chamberlin), 1 male, 1 female (JCC). ACKNOWLEDGMENTS We gratefully acknowledge the use of research facilities at the Malheur Field Station, near Burns, Oregon; the preparation of Figure 1 by Susan Linstedt of Eastern Oregon College; the loan of type specimens by John Unzicker of the Illinois Natural History Survey; and the loan of other valuable specimens by William B. Muchmore (WBM) of the University of Rochester. The holotype and allotype of the new species are deposited in the American Museum of Natural History (AMNH); the paratypes and other specimens are retained in the combined Benedict-Chamberlin-Malcolm Collection (EMB, JCC, and DRM) currently housed at Pacific University. LITERATURE CITED Banks, N. 1893. New Chernetidae from the United States. Canadian Entomol., 25:64-67. Banks, N. 1895. Notes on the Pseudoscorpionida. J. New York Entomol. Soc., 3:1-13. Beier, M. 1932. Pseudoscorpionidae. II Subord. Cheliferinea. Das Tierreich, 58:1-294. Beier, M. 1963. Ordnung Pseudoscorpionidae. Bestimmungsb. Bodenfauna Europas, 1:1-313. Benedict, E. M. 1978. A biogeographical study of currently identified Oregon pseudoscorpions with an emphasis on western Oregon forms. Ph.D. Dissertation. Portland State University, 123 pp. Chamberlin, J. C. 1924. Hesperochernes laurae, a new species of false scorpion from California inhabit- ing the nest of Vespa. Pan-Pacific Entomol., 1:89-92. Chamberlin, J. C. 1929. Dinocheirus tenoch, an hitherto undescribed genus and species of false scorpion from Mexico (Arachnida, Chelonethida). Pan-Pacific Entomol., 5:171-173. Chamberlin, J. C. 1931. The arachnid order Chelonethida. Stanford Univ. Pubis., Univ. Ser., Biol. Sci., 7:1-284. Chamberlin, J. C. 1934. On two species of false scorpions collected by birds in Montana, with notes on the genus Dinocheirus (Arachnida-Chelonethida). Pan-Pacific Entomol., 10:125-132. Chamberlin, J. C. 1952. New and little-known false scorpions (Arachnida, Chelonethida) from Monterey County, California. Bull. Amer. Mus. Nat. Hist., 99:259-312. Gering, R. L. 1956. Arachnids: spiders, pseudo scorpions, scorpions, solpugids. Pp. 49-50 in A. M. Woodbury, ed. Ecological check lists. The Great Salt Lake Desert Series. Univ. of Utah, Dugway. Hoff, C. C. 1944. New pseudoscorpions of the subfamily Lamprochernetinae. Amer. Mus. Novitates, 1271:1-12. Hoff, C. C. 1945. New species and records of pseudoscorpions from Arkansas. Trans. Amer. Micros. Soc., 64:34-57. Hoff, C. C. 1947. The species of the pseudo scorpion genus Chelanops described by Banks. Bull. Mus. Comp. Zool., 98:473-550. Hoff, C. C. 1949. The pseudoscorpions of Illinois. Bull. Illinois Nat. Hist. Survey, 24:407-498. BENEDICT AND MALCOLM-CHERNETID SPECIES FROM OREGON 109 Hoff, C. C. 1956. Pseudoscorpions of the family Chernetidae from New Mexico. Amer. Mus. Novitates, 1800:1-66. Hoff, C. C. 1957. Tejachemes (Arachnida-Chelonethida, Chernetidae-Chernetinae), a new genus of pseudoscorpion based on Dinocheirus stercoreus. Southwestern Nat. 2:83-88. Hoff, C. C. 1958. List of pseudo scorpions of North America north of Mexico. Amer. Mus. Novitates, 1875:1-50. Hoff, C. C. 1961. Pseudoscorpions from Colorado. Bull. Amer. Mus. Nat. Hist., 122:409-464. Hoff, C. C. and J. E. Bolsteri. 1956. Pseudoscorpions of the Mississippi River drainage basin area. Trans. Amer. Micros. Soc., 75:155-179. Hoff, C. C. and D. L. Clawson. 1952. Pseudoscorpions from rodent nests. Amer. Mus. Novitates, 1585:1-38. Knowlton, G. F. 1972. Some terrestrial arthropods of Curlew Valley. Utah State University Ecology Center, Terrestrial Arthropod Ser., 4:1-7. Lawson, J. E. 1968. Systematic studies of some pseudo scorpions (Arachnida: Pseudoscorpionida) from the southeastern United States. Ph.D. Dissertation. Virginia Polytechnic Institute. 297 pp. Legg, G. 1974a. A generalized account of the female genitalia and associated glands of pseudo- scorpions (Arachnida). Bull. British Arachnol. Soc., 3:42-48. Legg, G. 1974b. A generalized account of the male genitalia and associated glands of pseudoscorpions (Arachnida). Bull. British Arachnol. Soc., 3:66-74. Legg, G. 1975. Spermatophore formation in pseudoscorpions. Proc. 6th Int. Arachnol. Congr., pp. 141-144. Manley, G. V. 1969. A pictorial key and annotated list of Michigan pseudoscorpions (Arachnida: Pseudoscorpionida). Michigan Entomol., 2:2-13. Muchmore, W. B. 1971. A new Lamprochernes from Utah (Pseudoscorpionida, Chernetidae). Entomol. News., 82:327-329. Muchmore, W. B. 1974. Clarification of the genera Hesperochernes and Dinocheirus (Pseudoscor- pionida, Chernetidae). J. Arachnol., 2:25-36. Muchmore, W. B. 1975. Use of the spermathecae in the taxonomy of chernetid pseudo scorpions. Proc. 6th Int. Arachnol. Congr., pp. 17-20. Muchmore, W. B. 1976. Pseudoscorpions from Florida and the Caribbean area. 5. Americhemes, a new genus based upon Che lifer oblongus Say (Chernetidae). Florida Entomol. 59:151-163. Muchmore, W. B. 1981. New pseudo scorpion synonymies (Pseudoscorpionida, Chernetidae and Cheli- feridae). J. Arachnol., 9:258-260. Nelson, S. O., Jr., 1975. A systematic study of Michigan Pseudoscorpionida (Arachnida). Amer. Midi. Nat., 93:257-301. Nelson, S. O., Jr. and G. V. Manley. 1972. Dinocheirus horricus n. sp., a new species of pseudo- scorpion (Arachnida, Pseudoscorpionida, Chernetidae) from Michigan. Trans. Amer. Micros. Soc., 91:217-221. Weygoldt, P. 1966. Mating behavior and spermatophore morphology in the pseudoscorpion Dino- cheirus tumidus (Banks). Biol. Bull. 120:462-467. Weygoldt, P. 1970. Vergleichende Untersuchungen zur Fortpflanzungsbiologie der Pseudoscorpione II. Zeit. Zool. Syst. Evol. 8:214-259. Manuscript received October 1980, revised April 1981. Wise, D. H. 1982. Predation by a commensal spider, Argyrodes trigonum, upon its host: An experi- mental study. J. Arachnol., 10: 1 1 1-1 16. PREDATION BY A COMMENSAL SPIDER, ARGYRODES TRIGONUM , UPON ITS HOST: AN EXPERIMENTAL STUDY David H. Wise Department of Biological Sciences University of Maryland Baltimore County (UMBC) Catonsville, MD 21228 ABSTRACT Although the theridiid spider Argyrodes trigonum has beeh described as a commensal, it has been found eating its host. In central Maryland A. trigonum often inhabits the web of the labyrinth spider, Metepeira labyrinthea. I performed a field experiment to assess the possible impact of A. trigonum upon labyrinth spider populations. Replicated groups of M. labyrinthea were established on four open experimental units in the species’ natural habitat. Each unit was a wood frame supporting wire fencing on which the spiders built their webs. Mature A trigonum females were added to two of the units and numbers of the labyrinth spider were monitored for 18 days. Numbers of M. labyrinthea declined more rapidly on the units to which A. trigonum had been introduced. More dead labyrinth spiders were found in these populations, and indirect evidence suggests that losses from emigration may have been higher in the presence of A. trigonum. Labelling A. trigonum as a commensal is probably misleading. INTRODUCTION Spiders of the genus Argyrodes (Theridiidae) often behave as commensals, inhabitants of other species’ webs that consume prey apparently neglected or undetected by the host spider [Exline 1945, Archer 1946 (1947), Comstock 1948, Kaston 1948, 1978, Legendre 1960 (cited by Kaston 1965), Gertsch 1979]. However, some tropica \ Argyrodes spp. steal prey which the host has caught and thus clearly are kleptoparasites [Wiehle 1928, 1931, Thomas 1953, Kullmann 1959 (all cited by Kaston 1965), Robinson and Olazarri 1971, Robinson and Robinson 1973, Vollrath 1979]. Furthermore, some temperate Argyrodes spp. apparently prey upon their hosts. Exline (Exline and Levi 1962) has observed A. fictilium (Hentz) [= Rhomphaea lacerta (Walckenaer)] eating Araneus. Argyrodes fictilium will attack and eat Frontinella pyramitela (Walckenaer) [Archer 1946 (1947)], and Argyrodes spp. have been found consuming another linyphiid, the filmy dome spider, Neriene radiata (Walckenaer) (pers. obs.; J. Martyniuk, pers. comm.). Lamore (1958) observed the common Argyrodes trigonum (Hentz) [= Conopistha rufa (Walckenaer)] feeding upon a host basilica spider, Mecynogea lemniscata (Walckenaer). On several occasions my research assistants and I have found A. trigonum either feeding upon a basilica spider or in the web with the dead host. We have also observed A. trigonum eating the labyrinth spider , Metepeira labyrinthea Hentz. 112 THE JOURNAL OF ARACHNOLOGY I identified spiders as A. trigonum because of their similarity in size, color, and shape of the abdomen and egg sac, to the descriptions of this species given by Kaston (1948, 1978) and Exline and Levi (1962). Many Argyrodes that were found eating other spiders were not removed for a detailed examination because they were in the web of a host which was part of another ongoing field experiment. A. cancellatus (Hentz) is the only species with which A. trigonum occasionally may have been confused, particularly as juveniles. Most, if not all, spiders were probably correctly identified as A. trigonum. A. cancellatus may not be abundant in Maryland, since Kaston (1948) reports that it is quite rare in the north, but is common in Alabama. Also, Muma (1945) states that in Maryland A. trigonum occurs more often than A. cancellatus in the webs of M. labyrinthea. In some instances A. trigonum may have been feeding upon a host which had died from other causes. However, the frequency with which I have observed A. trigonum Fig. 1.— Effect of A. trigonum on numbers of M. labyrinthea. Population size of the labyrinth spider on each unit is expressed as the proportion of the number present on Day 0 (June 30). Numbers on all four units were similar at the beginning of the experiment. Initial population sizes on the control units were 27 and 31. On the units with A trigonum, initial numbers of M. labyrinthea were 30 and 32. WISE-PREDATION BY ARGYRODES TRICON UM UPON ITS HOST 113 eating other spider species suggests that it may regularly prey upon its host. To test this hypothesis I performed a field experiment to assess the potential impact of A. trigonum upon populations of M. labyrinthea. METHODS AND MATERIALS I conducted the experiment in mixed deciduous-pine woods on the Patuxent Wildlife Research Center, Prince Georges County, Maryland, U.S.A. This forest supports an abundant population of M. labyrinthea and has been the site of previous field experi- ments with this species (Wise 1979, 1981). Although labyrinth spiders spin webs on a variety of vegetation, for purposes of standardization the experiment was conducted with spiders that had spun webs on supports made of 5.1 cm mesh galvanized wire fencing (chicken wire) attached to wooden frames 4 m long, 2 m high, and 1 .6 m wide. The wire was arranged in an undulating pattern that ran the length of each unit. Two rows of 2 waves, each 1 m high, were separated by a 1.6 x 4 m horizontal piece of fencing, and a similar piece was also secured to the unit’s top. Four such structures ca. 10 m apart were used in the experiment. Units were not enclosed; thus no barriers prevented emigration and immigration of M. labyrinthea , their prey or their natural enemies. During the last week of June 1980, immature male and female labyrinth spiders were collected from the surrounding woods and added at random to empty experimental units. By 30 June 60% had constructed webs on the wire. Mature female A. trigonum with egg cases were collected from another site and on 30 June were added to two randomly selected units. I taped each A. trigonum's egg sac to the wire and then carefully placed the spider on her sac. Two units each received seven A. trigonum females, spaced evenly throughout the top half of each structure and as far as possible from occupied M. labyrinthea webs. Within two days most A. trigonum had detached the egg sac from the tape and had moved it to a newly constructed web 5-20 cm away from the point of introduction. No A. trigonum were added to the other two units, which served as con- trols. Neither M. labyrinthea nor A. trigonum individuals were marked. At the start of the experiment the numbers of M. labyrinthea in each population were: 27, 31 (control units) and 30, 32 (experimental units). Control and experimental populations were cen- sused ten times from 30 June through 18 July. Three A. trigonum , one of which was a male, colonized the controls. These immigrants were removed and were added to the units to which A. trigonum had been intentionally introduced. RESULTS AND DISCUSSION On 1 July, 12 of the 14 A. trigonum which had been added the previous day were either in their own web with an egg sac (8) or in the web of a labyrinth spider (4). After one week 1 1 A. trigonum were on the units, two in webs of M. labyrinthea. A week later eight were present, of which six occupied M. labyrinthea webs. By 18 July the number of A. trigonum dropped to three and I ended the experiment. The number of M. labyrinthea on the two units with A. trigonum declined rapidly during the first three days of the experiment, while numbers on the control units remain- ed stable (Fig. 1). On the third day the proportion of labyrinth spiders remaining in the two populations with A. trigonum was significantly lower than in the control populations [t= 11.1, p < 0.01 ; calculated from arcsin of the square root of the proportion, since proportions tend to be binomially distributed (Snedecor 1956)] . Between the third and 114 THE JOURNAL OF ARACHNOLOGY seventh day numbers in one control group declined suddenly: Nine of 27 spiders disap- peared from this unit, whereas only 3 of 31 disappeared from the other control. During this period an Argyrodes female colonized the first control unit and was found in a previously occupied labyrinth web on 7 July. It is questionable whether a single A. trigonum would cause so many host spiders to vanish, though it may have invaded several webs before it was discovered. For the next few days numbers of labyrinth spiders on the control units were again relatively stable. For the remainder of the experiment the aver- age number of labyrinth spiders in the experimental populations was lower than in the controls, though differences were no longer statistically significant as judged by t-tests. Plotting numbers on an arithmetic axis (as in Fig. 1), rather than logarithmic, reveals the magnitude of the initial drop in numbers, but makes it difficult to compare rates of population decline for the remainder of the study. The final number ofM labyrinthea on both experimental units, expressed as a proportion of the total number on the seventh day, was 0.28 (11/39). In the pooled control populations the proportion present at the end of the study was 0.67 (31/40). This difference suggests that A. trigonum was also affecting M. labyrinthea during the last half of the experiment; however, a statistical test was not employed since there was no a priori reason for calculating survival from the numbers present after a week. Since the units were unenclosed, the more rapid decline of M. labyrinthea in the experimental populations could have resulted solely from increased emigration in re- sponse to web invasions by A. trigonum. However, evidence indicates that predation by the commensal spiders contributed to the greater losses from the experimental popula- tions: On two occasions A. trigonum was observed eating its host, and twice A. trigonum was discovered in a web with a dead labyrinth spider. The overall documented mortality rate of M. labyrinthea was higher in the presence of A. trigonum. During the study 11 dead labyrinth spiders were found in their webs on the units to which A. trigonum had been added. Five of these spiders died during the first three days of the experiment. Only one dead spider was found among the control groups. The proportion of the decline in numbers during the experiment which was due to known deaths was 0.22 for the pooled experimental populations (1 1/51) and 0.04 for the combined control groups (1/27). This difference is statistically significant (p = 0.03, exact probability for the 2x2 contingency table). The decline in numbers for each treatment group, i.e. the net loss during the experiment, represents known mortality plus the net effects of immigration, emigration and undetected mortality. Some mortality in both populations may have resulted from other predators, such as the pirate spider, Mimetus puritanus Chamberlin. This specialized predator on other spiders occurred on control and experimental units during the experi- ment, and we have observed it preying upon M. labyrinthea at other times. Subtracting known mortality from the total decline in numbers yields the apparent net emigration, which includes undetected mortality. In order to compare control and experimental treatments, an apparent net emigration rate was calculated by expressing the number of apparent emigrants as a proportion of the total initial number after subtracting the number of known deaths. For example, the rate for the two pooled experimental populations was (51-1 l)/(62-l 1) = 0.78. Since some of this loss may have resulted from predators that remove their prey from the webs, comparing emigration rates calculated in this manner implicitly assumes that undetected mortality from sources other than A. trigonum was similar for all populations of M. labyrinthea. This assumption is also the basis for pooling replicates in order to calculate an overall emigration rate for the entire experiment. The apparent net emigration rate of the labyrinth spider was significantly higher on the units to which A. trigonum had been added (0.78 versus 0.46; WISE-PREDATION BY ARGYRODES TRIGONUM UPON ITS HOST 115 X2 = 12.2, p< 0.001, 2x2 contingency table), which suggests that some M. labyrinthea abandoned their webs in response to invasions by A. trigonum. Confirmation of this interpretation would require experimentation with marked labyrinth spiders on a larger number of experimental units. The initial rapid drop in numbers in the populations with A. trigonum, the greater number of dead spiders in these populations, and the observations of A. trigonum eating M. labyrinthea constitute evidence that the former species may be a significant mortality factor in the dynamics of labyrinth spider populations. An accurate assessment of the impact of A. trigonum upon its host populations will require further experiments. In particular, future studies should incorporate different life stages of both species as well as greater numbers of replicated populations. Describing Argyrodes spp. as commensals implies that they do not lower the fitness of their hosts. Accumulating evidence indicates that the relationship between Argyrodes spp. and their hosts is not always a benign commensalism. Although tropical Argyrodes species do capture small prey which apparently go unnoticed by the host, they also steal larger insects which the host has captured and wrapped. Several investigators have suggested that such kleptoparasitic behavior has a negative impact upon the host by causing it to move its web if the rate of prey stealing is too high (Robinson and Olazarri 1971, Robinson and Robinson 1976, Vollrath 1979). Rypstra (1981) has established a quan- titative relationship between number of Argyrodes in the host’s web, number of prey stolen, and tendency of the host, Nephila clavipes (L.), to abandon its web site. Energy requirements of constructing a new web and increased exposure to predation may lower the fitness of a spider which has vacated an otherwise suitable web in response to high rates of kleptoparasitism. No quantitative estimates exist on the effect of temperate Argyrodes spp. on the prey capture rates of their hosts. Perhaps the potentially competitive interactions between temperate Argyrodes spp. and their hosts are often more commensal than kleptoparasitic. However, generally describing such web-sharing spiders as commensals is incorrect, since the significant impact of A. trigonum upon the M. labyrinthea populations in this experi- ment suggests that the few reports of predation by temperate Argyrodes spp. actually reflect relatively frequent behavior. Trail [1980 (1981)] has found that A baboquivari is a conspicuous predator on the uloborid Philoponella oweni in the Chiricahua Mountains of Arizona, consuming eggs, hatchlings and adults. Most evidence of predation by Argyrodes spp. upon their hosts comes from temperate associations, though Lubin (1974) has discovered tropical Argyrodes kleptoparasites eating the eggs and recently emerged spiderlings of Cyrtophora moluccenis. Perhaps tropical kleptoparasites prey upon host species that are smaller than the conspicuous Nephila and Argiope spp. which have been studied extensively. Further examination of tropical Argyrodes associations should reveal whether this apparent behavioral difference is real. Valuable information would also be gained from further research on associations in temperate habitats. In particular, what is the role of factors such as host feeding rate, number of Argyrodes per web, and the relative size of host and Argyrodes in determining whether a particular interaction be- tween Argyrodes and its host is commensal, kleptoparasitic or predatory? The nature of the interaction may vary as a function of such variables. ACKNOWLEDGMENTS Field assistance was provided during 1980 by D. Bohle, K. Cangialosi, D. Debinski, J. Kirgan and D. Weiss. I also wish to acknowledge the contributions of the following field 116 THE JOURNAL OF ARACHNOLOGY assistants from previous years who observed Argyrodes spp. feeding upon their hosts: J. Barata, J. Capriolo, P. Ketzner, R. LaBelle and P. Pressman. I particularly wish to thank W. Stickel and the Patuxent Wildlife Research Center, U. S. Fish and Wildlife Service, for providing the undisturbed research site. W. Tietjen, J. Rovner and C. Horton made helpful suggestions on the manuscript. This research was supported by National Science Founda- tion Grant DEB 79-1 1744. LITERATURE CITED Archer, A. F. 1946 (1947). The Theridiidae or comb-footed spiders of Alabama. Alabama Mus. Nat. Hist. Paper No. 22., 67 p. Comstock, J. H. 1948. The spider book. Revised and edited by W. J. Gertsch. Ithaca, N. Y.: Comstock Publishing Co., 729 p. Exline, H. 1945. Spiders of the genus Conopistha (Theridiidae, Conopisthinae) from northwestern Peru and Ecuador. Ann. Entomol. Soc. Amer., 38:505-528. Exline, H. and H. W. Levi. 1962. American spiders of the genus Argyrodes (Araneae, Theridiidae). Bull. Harvard Mus.'Comp. Zool., 127:75-204. Gertsch, W. J. 1979. American spiders (Revised ed.) New York: Van Nostrand Reinhold Co., 274 p. Kaston, B. J. 1948. Spiders of Connecticut. Connecticut State Geol. and Nat. Hist. Survey Bull. No. 70, 874 p. Kaston. B. J. 1965. Some little known aspects of spider behavior. Amer. Midi. Nat., 73:336-356. Kaston, B. J. 1978. How to know the spiders (Third ed.). Dubuque, Iowa: Wm. C. Brown Co., 272 p. Kullmann, E. 1959. Beobachtungen und Betrachtungen zum Verhalten der Theridiide Conopistha argyrodes . Mitt. Zool. Mus. Berlin, 35: 275-292. Lamore, D. H. 1958. The jumping spider, Phidippus audax Hentz, and the spider Conopistha trigona Hentz as predators of the basilica spider, Allepeira lemniscata Walckenaer in Maryland. Proc. Ent. Soc. Washington. 60:286. Legendre, R. 1960. Quelques remarques sur le comportement des Argyrodes malgaches. Ann. Sci. Natur. Zool. ser 12, 2:507-512. Lubin, Y. D. 1974. Adaptive advantages and the evolution of colony formation in Cyrtophora (Araneae: Araneidae) Zool. J. Linn. Soc., 54:321-339. Muma, M. 1945. An annotated list of the spiders of Maryland. Maryland Agri. Exp. Stn. Bull. A-38, 65 P- Robinson, M. H. and J. Olazarri. 1971. Units of behavior and complex sequences in the predatory behavior of Argiope argentata (Fabricus) (Araneae: Araneidae). Smithson. Contrib. Zool., No. 65, 36 p. Robinson, M. H. and B. Robinson. 1973. Ecology and behavior of the giant wood spider Nephila maculata (Fabr.) in New Guinea. Smithson. Contr. Zool., No. 149, 76 p. Robinson, M. H. and B. Robinson. 1976. The ecology and behavior of Nephila maculata : a supple- ment. Smithson. Contrib. Zool., No. 218, 22 p. Rypstra, A. L. 1981. The effect of kleptoparasites on prey consumption and web relocation in a Peruvian population of the spider, Nephila clavipes (L.) (Araneae: Araneidae). Oikos, 37:179-182. Snedecor, G. W. 1956. Statistical methods. Iowa State University Press, Ames. 534 p. Thomas, M. 1953. Vie et moeurs des araignees. Payot, Paris. 339 p. Trail, D. S. 1980 (1981). Predation by Argyrodes (Theridiidae) on solitary and communal spiders. Psyche, 87:349-355. Vollrath, F. 1979. Behavior of the kleptoparasitic spider Argyrodes elevatus (Araneae: Theridiidae). Anim. Behav., 27:515-521. Wiehle, H. 1928. Beitrage zur Biologie der Araneen, insbesondere zur Kenntnis des Radnetzbaues. Z. Morph. Oekol. Tierre, 11:115-151. Wiehle, H. 1931. Neue Beitrage zur Kenntnis des Fanggewebes der Spinnen aus den Familien Argiopidae, Uloboridae und Theridiidae. Z. Morph. Oekol. Tierre, 22:349-400. Wise, D. H. 1979. Effects of an experimental increase in prey abundance upon the reproductive rates of two orb-weaving spider species (Araneae: Araneidae). Oecologia, 41:289-300. Wise, D. H. 1981. Inter- and intraspecific effects of density manipulations upon females of two orb-weaving spiders (Araneae: Araneidae). Oecologia, 48:252-256. Manuscript received January 1981, revised April 1981. Van Berkum, F. H. 1982. Natural history of a tropical, shrimp-eating spider (Pisauridae). J. Arachnol., 10:117-121. NATURAL HISTORY OF A TROPICAL, SHRIMP-EATING SPIDER (PISAURIDAE) Fredrica H. van Berkum Department of Zoology University of Washington Seattle, Washington 98195 ABSTRACT Trechalea magnified (Pisauridae) inhabit small streams in Costa Rica. Except for a few juveniles, these spiders were always active at night. This nocturnal activity may be due to temporal changes in predation risk and food availability. Predation risk was lower at night, whereas food was more abundant during the day. The diet of this species includes freshwater shrimp, a prey item previously unreported for spiders. These spiders construct hemispherical egg cases and carry them to their spinnerets: characteristics atypical of the family Pisauridae. INTRODUCTION The large pisaurid spider Trechalea magnifica Petrunkevitch [probably = T. extensa (0. P.-Cambridge), James E. Carico, pers. comm.] inhabits small streams in southwest Costa Rica. During the dry season (March 1980) in the Quebrada Camaronal (Corcovado National Park), adult and immature spiders were commonly active at night; but only a few small (immature) spiders were active in daylight. This activity pattern prompted an in- vestigation into why nocturnality might be advantageous to most of these spiders, and why some small spiders diverged from the usual pattern of activity. Possible reasons why most of these spiders were nocturnal include 1) food was more abundant at night and 2) predation pressure on the spiders was higher during the day. This study was designed to evaluate these possibilities. Some aspects of the natural history of the species are also included. In particular, I report the first case of shrimp-eating by a spider and a case of a pisaurid carrying its hemispherical egg sac attached to its spinnerets. MATERIALS AND METHODS Trechalea magnifica are well suited for study. They are large and did not flee when approached unless touched on the abdomen or cephalothorax. This behavior allowed me to mark spiders individually by touching a paint brush to unique locations on their legs. I counted T. magnifica resting on stones and logs in a 54 m2 area during the day and night to determine temporal patterns of spider activity. Spiders on substrates other than stones or logs were extremely hard to find during the day and thus were not censused at either census period. 118 THE JOURNAL OF ARACHNOLOGY To determine whether the temporal pattern of spider activity correlated with times of high food abundance, I measured the number of insects available to the spiders during the day and night by smearing square patches of tanglefoot (3 cm on a side) on two sets of twelve stones. One set was placed above the water line among stones in the stream for 10.5 hours at night, the other set was put out for the same amount of time and in the same place during the day. Trapped insects were counted and measured (lengths and widths) to the nearest millimeter. To estimate the relative abundance of shrimp during the day and night, I measured an index of shrimp activity. For twenty minutes during the day (0900 h) and night (2045 h), I counted the number of shrimp crossing a 1.1 m long by 3-4 mm wide piece of sub- merged cord. The cord was in the same place during the day and night. Only ‘prey-sized’ (small) shrimp were counted, and the night count was made using a flashlight covered with red plastic to minimize disturbance to the shrimp. Shrimp activity was thus a function of both shrimp numbers and movements. The influence of temporal activity pattern on risk of predation on spiders was deter- mined by tethering' 40 spiders, 20 during the day and 20 during the night, in their natural habitat. I tethered spiders by carefully tying threads around their pedicels and fastening the loose ends of the thread with tape to stones in the stream. Each group of 20 spiders (half were smaller than 1 cm in body length, half were larger) was left out for three hours (0800-1100 h or 1900-2200 h) after which the remaining individuals were counted. Tethered spiders rested quietly, as did their wild counterparts, since wild spiders did not try to escape until touched. Thus tethered spiders seemed to be reasonable mimics of wild spiders. I measured spider movements at night by observing 10 small (less than 1 cm in length) and 6 large (greater than 1 cm) individually marked spiders every 15 minutes from 1900 to 2200 h. I searched the same area for the marked spiders the following evening and noted changes in location. RESULTS Spider density.— During daytime censuses at 0700 h and 1400 h in the 54 m2 census plot, I found three and two small spiders, respectively. At night (2100 h) I counted 30 small and 18 large spiders. The densities of spiders during the two daytime censuses appeared lower than in some other parts of the stream. Throughout my study I never saw an adult spider active during the day. Prey availability and capture.— The number of prey sized insects trapped by tanglefoot during the day and night were similar (18 and 21, respectively) but the total volume of prey trapped [estimated by length times (width)2 , assumed to be proportional to bio- mass] was much higher during the day (118.85 mm3, day; 28.46 mm3, night). Even if the largest prey item trapped during the day (a 72 mm2 spider, the next to largest was 16 mm2) is excluded, the total prey volume trapped during the day was still twice as large as at night. Shrimp activity was nine crosses per 20 minutes at night, and 48 crosses per 20 minutes during the day. Consequently shrimp activity, and presumably availability to spiders, was higher during the day. Predation on spiders.— Of the 20 spiders tethered for three hours during the day, 16 were missing (seven large and nine small), two were being eaten by ants and two were uneaten. Of the 20 spiders tethered at night, one small spider was missing, three were VAN BERKUM-NATURAL HISTORY OF A TROPICAL PISAURID 119 being eaten by ants and 16 were uneaten (eight large and eight small). The 95% confi- dence intervals of the binomial probability of not being eaten during the day (2/20) and during the night (16/20) do not overlap. Thus predation pressure on T. magnified was considerably higher during the day, and the risk of predation appears equally severe for large and small spiders. Trechalea magnified appear to rely heavily on camouflage as a defense mechanism and do not respond to a disturbance around them until actually touched. Most vertebrate predators (birds, bats, lizards) should be able to instantly overpower even the largest spiders. Once a wild spider is found by a vertebrate predator, its chances of avoiding predation are limited. Therefore, although my predation data are most safely described as ‘rates of attack’ on spiders, they can also be interpreted as ‘rates of predation’. However, I did see spiders avoid ants in the field, so predation by ants was probably an artifact of the spiders being tethered. Excluding the spiders eaten by ants from the data does not affect the significance of the results. Natural history -The Quebrada Camaronal is a small, shallow stream running through second growth forest. Numerous ‘soft-ball sized’ stones (some larger, many smaller) fill the edge and shallow parts of the stream. Spiders rest primarily on these stones but also on logs, leaves, and the ground near the stream. The spiders hunt with bodies flattened and all eight legs spread more or less evenly to form a circle. Most large spiders place their first pair of legs on the water’s surface, although they can also be found in hunting position at some distance (generally never more than a meter) from the water. Some small spiders also hunt with their first legs on the water, but more commonly they are on top of stones that are in or near the water. This ontogenetic difference in hunting site probably reflects the inability of small spiders to capture the larger, active shrimp and probably in part caused a difference in the prey eaten by large and small spiders. Both large and small spiders fed on arthropods that flew or walked by, whereas large spiders also ate arthropods that floated on the water’s surface and aquatic prey. Williams (1979) reported that some pisaurids wait quietly for passing prey, whereas others dash out to capture more distant prey. T. magnifica seemed to follow the former strategy. I often saw spiders catch or attempt to catch insects attracted by my headlamp. Attacks consisted of quick, short lunges, usually no more than two leg-lengths away. Pisaurids do not require vision for prey capture, and can detect the buzzing of winged insects (Williams 1979). I saw at least half a dozen large spiders feed on the abundant fresh water shrimp in the stream. As far as I can determine, this is the first report of a spider that feeds on shrimp. Spiders have been reported to feed on birds (Bates 1876:83), mice and snakes (Gudger 1925), and isopods and amphipods (Lamoral 1968), in addition to insects and arachnids. Members of the family Pisauridae are well known for their ability to catch and eat fish (Gudger 1925, Williams 1979). I never saw the spiders feeding on fish, even though fish appeared as abundant as shrimp in the stream. Although I never saw a spider actually catch a shrimp in the field, a captive spider did capture a live shrimp (actual capture not observed). I often watched shrimp swim very near a hunting spider with no apparent response from the spider, therefore I assume that the spiders detect shrimp only when shrimp actually brush the spiders’ legs (Williams 1979). During the short time period of this study, the spiders appeared to be very site specific. On the night I monitored spider movements all of the spiders I watched re- mained within a one meter radius. Ten of the 16 spiders never left the rock on which they were resting, four of the remaining six moved only once, one moved twice, and a large 120 THE JOURNAL OF ARACHNOLOGY gravid female moved four times. The following night at 2015 I relocated nine of the 16 spiders. All but one large spider were within 10 cm of where I had observed them the night before. The large spider had moved about 5 m upstream. Although I did not search for spiders in their diurnal retreats, I assume for two reasons that they spend the day at some distance from their nocturnal hunting sites. First, I saw many large spiders hunting from rocks that seemed far too small for then to hide under. Second, many spiders hunted from stones that were partially submerged and surrounded by water. Although pisaurids often escape predation by submerging (Williams 1979), the spiders probably do not spend the day under water. These spiders seemed particularly incautious in the face of a large, potentially dan- gerous predator— myself. That they did not flee when touched on the legs probably explains the numerous spiders I saw with one to three legs missing. The large gravid female spider mentioned above was a conspicuous exception. She frequently dashed across the water when I approached, and would not allow me near enough to mark her. Also, spiders appeared to be more wary when they first emerged at sunset. Maternal behavior* in T. magnifica is unusual. Females construct hemispherical egg cases and carry them on their spinnerets. These properties are more characteristic of the closely related Lycosidae. In fact, pisaurids are often distinguished from lycosids by the round egg cases the pisaurids construct and carry with their chelicerae and pedipalps (Comstock 1913:602, Kaston 1978). T. magnifica are atypical of the family in this respect. DISCUSSION Nocturnal activity at exposed sites by Trechalea magnifica was not associated with higher food levels. In fact the opposite was true: food— both shrimp and other arthropods— was more abundant during the day. In contrast, nocturnal activity resulted in a far lower predation risk than diurnal activity. Consequently, nocturnal activity was less advantageous for food procurement, but highly advantageous for predator avoidance. I observed several potential predators during the daytime near the stream. Insecti- vorous birds and basilisc lizards were quite abundant, and large wasps were not uncom- mon. Carico (1973) reported that pompilid and sphecoid wasps are predators on Dolomedes tenebrosus (Pisauridae) in temperate areas. The low rate of predation on spiders during the night was unexpected because there were numerous potential nocturnal predators. I saw mygalomorph and lycosid spiders at night and bats in the gleaning carnivore guild were probably common in the area. Bonaccorso (1979) reported that these bats were common on Barro Colorado Island, Panama, an island that is climatically similar to Corcovado National Park. Gleaning carni- vores pick their prey off leaves and the ground. Perhaps the flattened profile assumed by T. magnifica while hunting allows them to avoid detection by a bat’s sonar. Why are some small spiders diurnally active despite high predation risks? My data do not provide a clear resolution, but suggest the following hypothesis. The scarcity of food at night may be more severe for small spiders because large spiders can eat shrimp as well as other arthropods. Perhaps food levels were so low for the small spiders that they risk high predation during the day to take advantage of the higher diurnal insect abundance. Small spiders might have been able to use small inconspicuous sites for hunting that were not available to large or tethered spiders. Thus I suspect that predation was not as severe on small spiders during the day as my data suggest. VAN BERKUM-NATURAL HISTORY OF A TROPICAL PISAURID 121 ACKNOWLEDGMENTS Special thanks to Chris Simon for help and support during nocturnal forays into caiman infested streams, and to Carlos Valerio and Raymond B. Huey for help with the manuscript. I am very gratful to James E. Carico for identifying the spiders and for his generous help. The data were collected while participating in the Organization for Tropical Studies 80-1 course. LITERATURE CITED Bates, H. B. 1876. The naturalist on the river Amazons, fourth edition. John Murray, London. Reprinted by Dover, New York, 1975. 394 pp. Bonaccorso, F. J. 1979. Foraging and reproductive ecology in a Panamanian bat community. Bull. Florida St. Mus. Biol. Sci., 24:359-408. Carico, J. E. 1973. The nearctic species of the genus Dolomedes (Araneae: Pisauridae). Bull. Mus. Comp. Zool., 144:435-488. Comstock, J. H. 1913. The spider book. Doubleday, Page and Co., New York. 721 pp. Gudger, E. W. 1925. Spiders as fishermen and hunters. Natural History, 25:261-275. Kaston, B. J. 1978. How to know the spiders, third edition. Wm. C. Brown Co., Dubuque. 272 pp. Lamoral, B. H. 1968. On the ecology and habitat adaptations of two intertidal spiders, Desis formi- dabilis (O. P.-Cambridge) and Amaurobioides africanus Hewitt, at “The Island” (Kommetjie, Cape Peninsula), with notes on the occurrence of two other spiders. Ann. Natal. Mus., 20:151-193. Williams, D. S. 1979. The feeding behaviour of New Zealand Dolomedes species (Araneae: Pisauridae). New Zealand J. Zool., 6:95-105. Manuscript received March 1981, revised May 1981. Maury, E. A. 1982. Solifugos de Columbia y Venezuela (Solifugae, Ammotrechidae). J. Arachnol., 10:123-143. SOLIFUGOS DE COLOMBIA Y VENEZUELA (SOLIFUGAE, AMMOTRECHIDAE) Emilio A. Maury Museo Argentino de Ciencias Naturales Angel Gallardo 470 (1405) Buenos Aires, Argentina ABSTRACT A systematic revision of the solifugids of Colombia and Venezuela is presented. All the species mentioned here belong to the family Ammotrechidae. Two. genera and one species are described as new: Xenotrecha, new genus, for X. huebneri (Kraepelin 1899) n. comb., and Eutrecha longirostris, new genus and new species. Two synonymies are proposed: Gluvia martha Karsch 1879 - Ammo- trechella geniculata (C. L. Koch 1842) and Ammotrechula vogli Roewer 1952 = Ammotrechella geniculata (C. L. Koch 1842); Mummuciona marcuzzii Caporiacco 1951 is tentatively considered as a synonym of M. simoni Roewer 1934. “Gluvia gracilis” C. L. Koch 1842 is definitively eliminated from the New World solifugid fauna. The other species reported in this contribution are Saronomus capensis (Kraepelin 1899) and Ammotrechula sp. Some subfamilial and generic characters employed in the systematics of the family Ammotrechidae are discussed. RESUMEN Se presenta una revision sistematica de los solifugos de Colombia y Venezuela. Todas las especies aquf mencionadas pertenecen a la familia Ammotrechidae. Se describen dos generos y una especie nuevos: Xenotrecha, genero nuevo, para X. huebneri (Kraepelin 1899) n. comb., y Eutrecha longi- rostris, genero y especie nuevos. Se proponen dos sinonimias: Gluvia martha Karsch 1879 = Ammo- trechella geniculata (C. L. Koch 1842) y Ammotrechula vogli Roewer 1951 = Ammotrechella geniculata (C. L. Koch 1842); Mummuciona marcuzzi Caporiacco 1951 es tentativamente considerada como un sinonimo de M. simoni Roewer 1934. “Gluvia gracilis” C. L. Koch 1842 es definitivamente eliminada de la fauna de solifugos del Nuevo Mundo. Las otras especies de solifugos mencionadas en esta contribution son Saronomus capensis (Kraepelin 1899) y Ammotrechula sp. Se discuten algunos caracteres empleados en la sistematica de la familia Ammotrechidae. INTRODUCCION La literatura referente a la fauna de solifugos del extremo norte de Sudamerica es sumamente escasa y consiste casi exclusivamente en las descripciones originales de unas pocas especies. Esta parquedad se debe principalmente a la falta de colectas adecuadas y sobre todo a la escasez de especialistas en este grupo de aracnidos, pues es indudable que las colecciones las forman los interesados. Es logico suponer que los vastos sistemas orograficos y las limitadas pero interesantes zonas xerofilas de esta parte de nuestro continente reunen condiciones muy propicias para la vida de los solifugos, tal como 124 THE JOURNAL OF ARACHNOLOGY sucede en regiones similares de America. Por esta razon supongo que el numero de formas alii existentes sera mucho mas elevado que las que menciono en esta contribucion. En este trabajo me ocupare exclusivamente de los solifugos de Colombia y Venezuela. Las especies de otras regiones cercanas, como Centroamerica y las Antillas fueron trata- das, algo sucintamente, por Muma (1970). Como dato inedito, y para completar el cono- cimiento de la distribucion de estos aracnidos por el norte de Sudamerica, mencionare la presencia de solifugos en Surinam y en el estado de Roraima, Brasil. Segun la literatura consultada, para Colombia y Venezuela han sido mencionadas las siguientes especies de solifugos (se citan con la grafia original): Gluvia geniculata C. L. Koch 1842, Gluvia gracilis C. L. Koch 1842, Gluvia martha Karsch 1879, Cleobis gervaisi Pocock 1895, Sarophorus capensis Kraepelin 1899, Cleobis hubneri Kraepelin 1899, Mummuciona simoni Roewer 1934, Mummuciona marcuzzii Caporiacco 1951 y Ammotrechula vogli Roewer 1952. Las modificaciones nomenclatoriales que han sufrido varias de esas especies seran tratadas en detalle mas adelante. Respecto a “Gluvia gracilis”, descripta supuestamente de Colombia, el examen del tipo demuestra que se trata de un solifugo africano, tal como suponia Kraepelin (1901:21). En lo referente a “Cleobis gervaisi ”, especie de localidad tipica algo dudosa, el mismo Pocock menciona ejemplares de Colombia, pero lamentablemente ese material se ha perdido. El estudio de los solifugos de Colombia y Venezuela me ha enfrentado a singulares problemas taxonomicos, que atanen no solo a las especies de esta region sino en general a toda la familia Ammotrechidae. Dos caracteres morfologicos que desde Roewer (1934) en adelante han merecido la aprobacion o la critica de numerosos especialistas necesitan de una mencion especial. Son ellos la subdivision de los tarsos de las patas II, III y IV (caracter primordial segun Roewer para distinguir las subfamilias de Ammotrechidae) y la espinulacion de esos mismos tarsos (caracter que, siempre segun Roewer, permite la separacion generica dentro de las mencionadas subfamilias). A continuacion hare unas breves consideraciones sobre la utilidad y las limitaciones de ambos caracteres en este estudio sistematico. En los solifugos, los tarsos de las patas II, III y IV pueden o no estar divididos en subsegmentos. En otros grupo de artropodos con tarso subdividido (vgr. Insecta, Opiliones) cada uno de estos subsegmentos recibe el nombre de tarsito (o tarsomero). Creo apropiado emplear tambien en Solifugae el termino tarsito, lo que uniformizara la nomenclatura y evitara el empleo de palabras de signification algo ambigua (en castellano se han usado los terminos artejo, articulo, eslabon). En los Ammotrechidae he compro- bado de que existe por lo menos un genero: Oltacola, en el cual la subdivision tarsal es bien evidente y en donde cada tarsito aparece independiente y claramente diferenciado de los adyacentes. Esto mismo se observa en varias familias de solifugos, Solpugidae es quiza el caso mas notable, y tambien algunos Daesiidae como Biton, Eberlanzia, etc. En los restantes Ammotrechidae la segmentation tarsal es mucho menos manifiesta y esto ha sido la causa de malinterpretaciones y errores al emplearla como caracter taxonomico. Como la visualization de este caracter, aun empleando considerables aumentos, puede presentar ciertas dificultades, considere apropiado ensayar dos metodos tradicionales en el estudio de la quitina de los artropodos. Algunos tarsos seleccionados se trataron con hidroxido de sodio al 10% en caliente, con el objeto de eliminar las partes blandas y aflojar la cobertura pilosa. Otros tarsos se colorearon con fuscina acida de Gage al 10% por espacio de 12 horas, seguido de un lavado de 24 horas en agua destilada para eliminar el exceso de colorante. Ambos metodos me permitieron comprobar (me refiero solamente a los solifugos estudiados en este trabajo) que la tenue separacion entre tarsito y tarsito MAURY- SOLIFUGOS DE COLOMBIA Y VENEZUELA 125 corresponde simplemente a un adelgazamiento de la quitina en un sector anular, lo que da a esa zona una mayor flexibilidad. Dicha zona esta desprovista de pelo, y con la fuscina se colorea mas debilmente que el resto del tarso. Un dato interesante a mencionar es el que los machos, posiblemente por un alargamiento del tarso, presentan la subdivision en tarsito mas evidente que las hembras. Esto es particularmente notable enAmmotrechella geniculata. El desprendimiento de la cobertura pilosa y la coloracion trajeron aparejado otro hallazgo: en el extremo distal de los tarsos II, III y IV existe una plaquita ventral, de forma aproximadamente triangular cuando se la examina de lateral y separada del resto del tarso por un surco bien manifesto. Aparentemente esta plaquita no fue observada por ningun investigador de los varios que se ocuparon de la morfologia de los solffugos, pues en la literatura consultada no encontre que la mencionaran. En algunas especies, como Ammotrechella geniculata , y especialmente en los tarsos II y III la plaquita no es facil de ver, sobretodo por estar oculta tras la abundante cobertura pilosa; pero en otras, como Xenotrecha huebneri es bien evidente (Fig. 26). Considero a esta plaquita como un tarsito mas, y de indudable importancia taxonomica ya que puede o no llevar espinas. Este hallazgo trae como consecuencia la necesidad de un redimensionamiento del caracter “subdivision tarsal”, que indudablemente modificara los actuales conceptos subfamiliares en Ammotrechidae. Aunque el tarsito terminal mencionado fue observado en todos los solffugos estudiados en este trabajo, por el momento no estoy en condiciones de aflrmar que el mismo se encuentre presente en todos los Ammotrechidae. Respecto al otro caracter taxonomico en discusion (la espinulacion tarsal), tambien la coloracion con fuscina me fue de cierta ayuda, ya que las espinas, quizas por su mayor quitinizacion, se colorean de una manera algo diferente a la de los pelos de cobertura. Roewer dio una tremenda importancia a la constancia de la formula espinular de los tarsos, tanto es asf que algunas de sus claves de diagnosis genericas parecen en realidad tablas de formulas matematicas. La fijeza de este caracter ha sido discutida mas de una vez, y el presente estudio corrobora este hecho: en cuatro de los seis generos mencionados en este trabajo he comprobado una llamativa variabilidad en el numero de espinas tarsales. En todos los casos esta variabilidad afecto exclusivamente las espinas del ultimo tarso, en las patas II, III y IV, y consistio siempre en una espina de diferencia con la formula espinular “normal”. Halle esta variabilidad en diferentes ejemplares de una misma especie y aun en un mismo animal, si tomamos en consideration el lado derecho o el izquierdo del especimen en cuestion. Ver, por ejemplo, lo manifestado respecto al unico ejemplar estudiado de Eutrecha longirostris. Es indudable que frente a esta variabilidad, el caracter “espinulacion tarsal” no debe ser empleado con un valor tan categorico y definitorio como pretendfa Roewer. Por esta razon considere apropiado mencionar las variaciones que encontre en cada uno de los tarsos en las diagnosis genericas que doy mas adelante. Cuando es posible (en mi caso, solo con Ammotrechella geniculata), senalo el porcentaje de ejemplares que posefan una formula espinular “normal” 6 “ideal”, es decir la presente en la mayorfa de los ejemplares. Se notara tambien que tres generos: Xenotrecha, Mum- muciona y Eutrecha pueden presentar en los tarsos II y III la misma formula espinular: 2. 2. 2/ 1.1, pero esto no es obice para que una buena proportion de otros caracteres diferenciales me permita afirmar que se trata de generos distintos. Al realizar este trabajo surgio la conveniencia de incluir en las diagnosis genericas de los Ammotrechidae algunos caracteres mas que los que habitualmente se emplean. En este caso, ademas de la segment- ation y espinulacion tarsales, he comprobado el valor que pueden tener la espinulacion de los protarsos de las patas II y III; la espinulacion de los pedipalpos; la presencia, distri- bution y caracterfsticas de los ctenidios en los esternitos espiraculares I y II (en el macho 126 THE JOURNAL OF ARACHNOLOGY exclusivamente); el operculo genital (en la hembra) y por supuesto, las caractensticas de la dentition de los queliceros en ambos sexos y la del flagelo en el macho. En las claves de identification de las especies mencionadas en este trabajo he tratado de utilizar en lo posible los caracteres mas facilmente abordables y solo en ultima instancia empleo la segmentation o espinulacion tarsal. Parece incuestionable que la diferenciacion subfamiliar de los solifugos Ammo- trechidae debera ser en gran parte modificada y actualizada a la luz de los caracteres que acabo de mencionar. Pero esta es una labor compleja, que necesitara forzosamente del estudio de todo el material tfpico, pues es evidente que las descripciones originales, al omitir muchos datos y tergiversar otros, puede llevar al engano. CLAVE PARA LAS ESPECIES MACHOS 1. Presencia de ctenidios en el I y II esternitos espiraculares (Figs. 23, 40); protarsos de los pedipalpos con 8 a 12 pares de espinas ventrales (Figs. 24, 41) 2 Presencia de ctenidios solo en el I esternito espiracular (Figs. 3, 15,31); protarso de los pedipalpos con 3 a 5 pares de espinas ventrales (Figs. 4, 32) . 3 2. Numerosos ctenidios en cada esternito (Fig. 23); dedo movil de los queliceros con un diente basal interno, dedo fijo con cuatro dientes basales internos (Fig. 22) . . . Xenotrecha huebneri Pocos ctenidios (6-10) en cada esternito (Fig. 40); dedo movil de los queliceros sin diente basal interno, dedo fijo con tres dientes basales internos (Fig. 39) Mummuciona simoni 3. Dedo fijo de los queliceros sin dientes anteriores (Fig. 29); tibia de los pedipalpos con 4 pares de espinas ventrales (Fig. 32) Eutrecha longirostris Dedo fijo de los queliceros con dientes anteriores (Figs. 1, 13); tibia de los pedipalpos sin espinas 6 con 1.1.1 espinas ventromedianas (Fig. 4) 4 4. Borde dorsal de dedo fijo de los queliceros con un grupo de setas modificadas, hay 3 dientes anteriores (Fig. 1); tibia de los pedipalpos con 1.1.1 espinas (Fig. 4) . . . . Saronomus capensis Borde dorsal del dedo fijo de los queliceros sin setas modificadas, hay 2 dientes anteriores (Fig. 13); tibia de los pedipalpos sin espinas ventrales Ammotrechella geniculata HEMBRAS 1. Dedo movil de los queliceros sin diente basal interno; dedo fijo con 3 dientes basales internos (habitualmente falta el 2- diente, si esta presente es vestigial) (Fig. 36) . . Mummuciona simoni Dedo movil de los queliceros con un diente basal interno; dedo fijo con 4 dientes basales internos (Figs. 10, 19) 2 MAURY- SOLIFUGOS DE COLOMBIA Y VENEZUELA 127 2. Protarso de los pedipalpos con 3 a 4 pares de espinas ventrales 3 Protarso de los pedipalpos con 7 a 10 pares de espinas ventrales 4 3. Tarso IV con 3 tarsitos; espinulacion del tarso III: 2.2.2/2.1, del tarso IV: 2. 2.2/2/ 1 . Saronomus capensis Tarso IV con 4 tarsitos; espinulacion del tarso III: 1.2. 2/1 6 1.2. 2/2, del tarso IV: 2.2/2/2/0 6 2.2/2/2/1 Ammotrechella geniculata 4. Tibia de los pedipalpos sin espinas ventrales; tarso IV con 3 tarsitos; espinulacion del tarso III: 2.2.2/1 6 2.2.2/1.1, del tarso IV: 2.2.2/2/0 Xenotrecha huebneri Tibia de los pedipalpos con espinas ventrales; tarso IV con 4 tarsitos; espinulacion del tarso III: 1. 2.2/2. 1, del tarso IV: 2.2/212/2 Ammotrechula sp. Saronomus Kraepelin Sarophorus Kraepelin 1899b: 234 (no Sarophorus Erichson 1$47). Saronomus Kraepelin 1900: 7, 1901: 106; Roewer 1934: 580, 1941: 178; Mello-Leitao 1938a: 11, 1938b: 266; Muma 1976: 22. Especie tipo '.-Saronomus capensis (Kraepelin 1899) por monotipia. Distribution: —Venezuela, Colombia. Diagnosis: —Ammo t rechid ae con los tarsos II y III con 2 segmentos y el tarso IV con 3 segmentos. Espinulacion de los tarsos II y III: 2. 2. 2/2.1; espinulacion del tarso IV: 2. 2.2/2/ 1. Dedo movil de los queliceros con diente basal interno. Macho con ctenidios en el esternito espiracular I. Macho con un grupo de setas cilmdricas de extremo crateri- forme en el borde dorsal de los queliceros. Macho con tres dientes anteriores en el dedo fijo de los queliceros; hembra con dos dientes. Saronomus capensis (Kraepelin 1 899) (Figs. 1-8) Sarophorus capensis Kraepelin 1899: 235. Saronomus capensis Kraepelin 1900: 7, 1901: 107; Roewer 1934: 581; Goetsch und Lawatsch 1944: 80; Caporiacco 1951: 35; Muma 1976: 22. Description del holotypus macho.- Medidas en milimetros: Tabla I. Coloration: ejemplar destenido por la fijacion:. color castano claro uniforme. Morfologia: Prosoma: propeltidio un poco mas ancho que largo (mdice|largo/ancho: 0,96). Lobulos laterales poco prominentes, separados del propeltidio por un surco dorsal. Tuberculo ocular con los ojos separados algo menos de un diametro ocular. Todo el propeltidio cubierto de pequenas setas, entre las que se destacan algunas dispersas mucho mas largas y gruesas. Los otros segmentos del prosoma y los tergitos con el mismo tipo de pelos de cobertura. Esternitos cubiertos de finas setas terminadas en furcula; esternito espiracular I con dos densas areas de ctenidios (Fig. 3). Queliceros (Figs. 1-2): dedo movil con el mucron largo, aguzado, sin curvatura dorsal. Dentition: diente anterior y diente intermedio de tamano similar, el diente principal algo mas grande, los tres separados por amplias diastemas. 128 THE JOURNAL OF ARACHNOLOGY Figs. 1-6. -Saronomus capensis (Kraepelin), holotypus macho: 1, quelfcero derecho, vista externa; 2, quelfcero derecho, vista interna; 3, operculo genital y esternitos espiraculares I y II (semiesque- matico); 4, pedipalpo derecho, vista interna; 5, tarso II derecho, vista ventral; 6, tarso IV derecho, vista lateral. Figs. 7-8.- Saronomus capensis (Kraepelin), hembra: 7, quelfcero derecho, vista externa; 8, oper- culo genital. MAURY— SOLIFUGOS DE COLOMBIA Y VENEZUELA 129 Tabla l.-Medidas en milimetros de los ejemplares descriptos Saronomus capensis Ammotrechella geniculata Xenotrecha huebneri Mummuciona simoni Eutrecha longirostris 6 Hoi. 9 9 Syn. d 9 Hoi. d 9 Syn. d d Hoi. Longitud total 24,96 21,44 17,28 17,92 14,40 17,15 11,90 8,96 13,63 Quelicero, longitud 4,99 6,40 5,89 4,22 2,75 3,71 2,56 2,56 3,14 Quelicero, alto 1,60 2,37 1,92 1,28 0,96 1,22 0,83 0,70 0,87 Propeltidio, longitud 3,33 3,52 3,78 2,81 2,18 2,94 1,79 1,73 2,43 Propeltidio, ancho 3,46 5,38 4,48 3,14 2,50 3,07 1,98 1,98 2,43 Pedipalpo, longitud 17,54 14,08 11,84 13,37 7,10 12,92 6,14 11,90 11,84 Pata I, longitud 12,60 10,43 8,06 8,70 4,35 6,20 3,90 5,31 6,46 Pata IV, longitud 20,48 16,64 — 14,91 7,68 13,28 6,66 11,58 12,16 Dedo fijo de borde dorsal sinuoso, con una concavidad a la altura de los dientes ante- riores. Mucron corto, encorvado y algo inclinado hacia lateral. Dentition: tres dientes anteriores pequenos, equidistantes y de tamano similar, le sigue un diente intermedio ligeramente mayor que los anteriores, luego, separado por una amplia diastema, el diente principal y a continuation cuatro dientes basales internos, que en orden decreciente de tamano se ordenan: 3-1-4-2, hay cuatro dientes basales internos, 1 y 3 largos y aguzados, 2 pequeno y 4 unido por la base al 3. Borde dorsal del dedo fijo, a la altura de la articulation del dedo movil, con un grupo de 9-10 robustas setas cilmdricas, de extremo crateriforme y dilatado. El flagelo sobrepasa en casi toda su extension el borde dorsal del quelicero, se extiende desde el tercio anterior del mucron hasta la altura del cuarto diente basal interne, tiene una forma piriforme alargada con el extremo distal ornado de dimi- nutas espiculas. Anillo de fijacion eliptico, situado a la altura del primer diente basal externo. Pedipalpos (Fig. 4): protarso con 2.2.2 robustas espinas; tibia con 1.1.1 espinas ventromedianas. Patas: protarso de la pata II con 1.1.1 espinas dorsoposteriores y 1.1.2 espinas ventrales; protarso de la pata III con 1.1.1 espinas dorsoposteriores y 1.1. 1.2 ventrales; protarso de la pata IV con 1.1. 1.2 espinas ventrales. Tarsos (Figs. 5-6) II y III con dos segmentos y con la espinulacion: 2.2.2/2.1; tarso IV con tres segmentos y con la espinulacion: 2.2.2/2/1. Description de un ejemplar hembra (Ipapure, Colombia). -Medidas en milimetros: Tabla I. Coloration: prosoma y tergitos con reticulado castano oscuro sobre un fondo amarillento; faz dorsal de los pedipalpos y patas con un leve tinte castano oscuro; el resto del ejemplar color amarillento. Morfologia: propeltidio bastante mas ancho que largo (indice largo/ancho: 0,66). Queh'ceros (Fig. 7) sin el grupo de setas cilmdricas dorsales; borde dorsal del dedo fijo con una prominencia a nivel del diente principal. Dedo movil con un diente anterior, un diente intermedio pequeno, un diente principal robusto y un diente basal interno bien desarrollado. Dedo fijo con dos dientes anteriores de tamano similar, un intermedio mas pequeno, un principal que es el mayor de la serie, cuatro basales externos que en orden decreciente de tamano se ordenan: 1 -3-4-2 y cuatro basales internos bien desarrollados. Primer esternito espiracular sin ctenidios. Pedipalpos: protarso con 2.2.2 espinas; tibia sin espinas. Operculo genital como se indica en la Fig. 8. Variabilidad.— Debido al escaso material que he podido estudiar de esta especie (2d, 19), poco es lo que puedo decir de la variation interespecifica. El macho proveniente de Colombia es algo mas pequeno (mide 17 mm de longitud total) que el holotypus, y las 130 THE JOURNAL OF ARACHNOLOGY setas cilmdricas agrupadas en el borde dorsal del quelicero estan en menor numero (5-6). Otros caracteres, como la dentition de los queliceros y la espinulacion de patas y pedi- palpos son similares en los dos ejemplares. Comentarios.— En la description original de “Sarophorus capensis” Kraepelin (1899) incurre en dos errores: utiliza el nombre generico Sarophorus , preocupado en Coleoptera por Erichson (1847) e indica como localidad tipica “Capland” (El Cabo, Sudafrica). Alertado por colegas, Kraepelin prontamente corrige esos errores y en 1900 emplea la nueva denomination Saronomus capensis y senala la verdadera localida tipica: Peninsula de Paraguana, en Venezuela. Kraepelin (1899, 1901) ubicaba a Saronomus en la sub- familia Daesiinae (familia Solpugidae). En la revision mundial del Orden Solifugae efec- tuada por Roewer (1932-34), este autor incluye a Saronomus en la familia Ammo- trechidae y lo considera el genero tipo de la subfamilia Saronominae. Desde 1934 fueron incluidos en dicha subfamilia siete generos (Muma 1976) pero posteriormente Maury (1977) efectua algunas correcciones, eliminando tres generos de Mello-Leitao e incor- porando un genero de Kraepelin. Saronominae fue definida por Roewer (1934) como “Ammotrechidae cuyos tarsos II, III y IV presentan siempre un solo artejo”. Pero como he mencionado en la introduction, un examen atento de los tarsos de los solifugos estudiados me ha revelado que no siempre las subdivisiones tarsales mencionadas por Roewer eran las reales. En el caso de Saronomus en donde los tarsos II y III poseen dos segmentos (tarsitos) y el tarso IV tres segmentos, esta subdivision lo apartaria de lo que podria considerarse un “tipico” Saronominae. Pero como Saronomus es el genero tipo de esta subfamilia, es evidente que se impone una redefinition de Saronominae y una reubicacion de los generos que se le atribuyen. Dada la indole del presente trabajo y considerando que Saronominae no es el unico problema referente a la validez de las subfamilias de Ammotrechidae (ver lo expresado respecto a Mummuciinae), dejare este tema para una futura contribution. Especimenes estudiados. -VENEZUELA: Peninsula de Paranagua (sin fecha ni colector), holotypus macho (MNHN 105). COLOMBIA: DepartamentoGuajira, Merochon, 5 Km de Uribia, 2-3 de septiem- bre de 1969 (B. Malkin), un macho (AMNH); Ipapure, 22-23 de septiembre de 1968) (B. Malkin), una hembra (AMNH). Ammotrechella Roewer Ammotrechella Roewer 1934: 594, 1941: 182; Mello-Leitao 1938a: 22; Muma 1951: 125, 1970: 45, 1976: 25; Muma and Nazario 1971 : 506. Especie tipo —Ammotrechella geniculata (C. L. Koch 1842) por designation original. Distribution.— Venezuela, Colombia. Tambien citada para Ecuador, Cura9§o, Trini- dad, Bahamas, Guadeloupe, Saint Vincent. Diagnosis.— Ammotrechidae con los tarsos II y III con dos segmentos y el tarso IV con cuatro segmentos. Espinulacion de los tarsos II y III: 1.2. 2/1 6 1.2. 2/2; espinulacion del tarso IV: 2.2/2/2/0 6 2.2/2/2/1. Dedo movil de los queliceros con diente basal interno. Macho con ctenidios en el esternito espiracular I. En ambos sexos dos dientes anteriores en el dedo fijo de los queliceros. Ammotrechella geniculata (C. L. Koch 1842) (Figs. 9-17) Gluvia geniculata C. L. Koch 1842: 355, 1848: 98; Butler 1873: 424 (en parte); Karsch 1880: 232. Ammotrecha geniculata: Kraepelin 1901: 114. MAURY- SOLIFUGOS DE COLOMBIA Y VENEZUELA 131 Ammotrechella geniculata : Roewer 1934: 594; Goetsch und Lawatsch 1944: 80; Zilch 1946: 151; Caporiacco 1951: 35;Muma 1970: 46, 1976: 25;Mumaand Nazario 1971: 507. Gluvia martha Karsch 1879: 108; Roewer 1934: 606. NUEVA SINONIMIA. Cleobis martha: Karsch 1880: 237. Ammotrecha martha'. Kraepelin 1901: 115. Ammotrechula vogli Roewer 1952: 38. NUEVA SINONIMIA. Descripcion de un syntypus hembra.— Este ejemplar se encuentra conservado en seco, atravesado por un alfiler. Le faltan varias patas y el abdomen esta deformado, por lo que la longitud total que doy en la tabla es aproximada. Medidas en milfmetros: Tabla I. Coloracion: desvirtuada por la mala conservation. Morfologia: Prosoma: propeltidio algo mas ancho que largo (indice largo/ancho: 0,84). Lobulos laterales prominentes, separados del propeltidio por un surco dorsal. Tuberculo ocular con los ojos separados poco menos de un diametro ocular. Todo el propeltidio con setas dispersas de diferente grosor y largo. Los restantes segmentos del prosoma y los tergitos con el mismo tipo de pelos. Queliceros (Figs. 9-10): dedo movil con el mucron largo y con una leve curvatura dorsal. Denticion: un diente anterior y un diente principal de tamano similar, entre ellos un pequeno diente intermedio; el diente basal interno bien desarrollado. Borde inferior del dedo con una hilera de fuertes setas, que se continua en semicirculo por la cara interna. Dedo fijo: borde dorsal con una marcada prominencia a nivel del primer diente basal externo. Denticion: dos dientes anteriores de tamano similar, un diente intermedio pequeno, un diente principal robusto, cinco dientes basales externos de los cuales 1, 2 y 4 de tamano similar, el 3 un poco mas grande, el 5 vestigial, y cuatro dientes basales internos, 1 y 3 bien desarrollados, 2 y 4 mas pequenos. Pedipalpos (Fig. 12): cortos y robustos, protarso con 2.2. 2. 2 cortas y gruesas espinas, el par basal algo mas debil que los restantes; tibia sin espinas. Patas: protarsos II y III con 1.1 espinas dorsoposteroires y 1.1.2 espinas ven- trales. El protarso IV falta. Tarsos II y III con dos segmentos y con la espinulacion 1. 2.2/1. El tarso IV falta. Operculo genital como se indica en la Fig. 11. Descripcion de un ejemplar macho (Rancho Grande, Venezuela).— Medidas en mili- metros: Tabla I. Coloracion: prosoma, pedipalpos (especialmente tarso, protarso y tibia) y cara corsal de las patas color castano oscuro; tergitos con dos bandas longitudinales castano oscuro delimitando una banda mediana color amarillento; el resto del animal color amarillento. Morfologia: Prosoma: propeltidio un poco mas ancho que largo (mdice largo/ancho: 0,89). Lobulos laterales poco prominentes. Ojos separados algo menos de un diametro ocular. Esternitos: esternito espiracular I con dos areas de largo ctenidios (Fig. 15). Queliceros (Figs. 13-14) con largas setas dorsales y laterales. Dedo movil con una denticion similar a la del syntypus hembra, el mucron es ligeramente mas recto y sin la marcada curvatura dorsal; se ve tambien el arco de gruesas setas ventrointernas. Dedo fijo de borde dorsal casi recto, salvo una leve prominencia a nivel del diente principal. Mucron casi recto, levemente curvado en el extreme distal. Denticion similar a la del syntypus hembra, salvo que hay cuatro dientes basales externos. Flagelo ovoide alargado, sobrepasa ligeramente el borde dorsal del dedo fijo y se extiende desde la altura del primer diente anterior hasta algo por detras del cuarto diente basal interno; anillo de fijacion situado a nivel del segundo diente basal interno. Pedipalpos: protarso con 2.2.2 gruesas espinas, tibia sin espinas. Patas: espinulacion de los protarsos II y III similar a la del syntypus hembra; espinulacion del protarso IV: 1.1.2 espinas ventrales. Tarsos (Figs. 16-17) II y III con la espinulacion 1.2. 2/1; tarso IV con cuatro segmentos y con la espinulacion 2.2/2/2/0. Variabilidad.— Ammotrechella geniculata es la unica especie de solifugo de la region de la cual he podido estudiar suficiente material (49 ejemplares) como para hacer un breve 132 THE JOURNAL OF ARACHNOLOGY Figs. 9-12. Ammotr echella geniculata (C. L. Koch), syntypus hembra: 9, quelfcero derecho, vista externa; 10, quelfcero derecho, vista interna; 11, operculo genital; 12, pedipalpo derecho, vista in- terna. Figs. 13-17 -Ammotrechella geniculata (C. L. Koch), macho: 13, quelfcero derecho, vista externa; 14, quelfcero derecho, vista interna; 15, operculo genital y esternitos espiraculares I y II (semiesque- matico); 16, tarso III derecho, vista ventral; 17, tarso IV izquierdo, vista lateral. MAURY- SOLIFUGOS DE COLOMBIA Y VENEZUELA 133 comentario sobre la variabilidad de ciertos caracteres. Longitud total observada: hembras: entre 14 y 26 mm; machos: entre 14 y 19 mm. Espinulacion de los pedipalpos: Esta especie no posee espinas en la tibia. En el protarso, la espinulacion mas comun son cuatro pares de espinas: 2. 2. 2.2, pero como la serie dorsal es algo mas fuerte que la ventral y el par basal suele ser mas debil que los restantes, he podido observar las variaciones 2.2.2 6 1. 2.2.2. Las espinas que faltan de la serie normal no se diferencian facilmente de las setas de cobertura. Excepcionalmente he encontrado un ejemplar con 2.2. 2. 2.2 espinas. He observado que los pullus no present an espinas en los pedipalpos. Espinulacion los tarsos. La formula espinular mas comun (que denomino formula “normal” o “ideal”), presente en aproximadamente el 70% de los ejemplares estudiados, es la siguiente: tarsos II y III: 1 .2.2/1 ; tarso IV: 2.2/2/2/0. En los restantes ejemplares he hallado las siguientes va- riantes; tarsos II y III: 1.2. 2/2 y tarso IV: 2.2/2/2/1. Es de hacer notar que la diferencia con la espinulacion “normal” radica solamente en la adicion de una espina “suplemen- taria” en el ultimo segmento del tarso. Dicha espina parece ser siempre la terminal posterior. Comentarios.— La description original que C. L. Koch (1842) da para “Gluvia geni- culata” es brevisima, e indica como localidad tipica a Venezuela. En 1848 el mismo autor ofrece una redescripcion mas detallada y la lamina en color que la acompana es bien demostrativa, especialmente en lo referente al color de los tergitos y pedipalpos. En este trabajo la localidad tipica senalada es “Sudamerika, in den Gegenden des Orinoko” (Sudamerica, en la region del Orinoco). Citaciones posteriores, especialmente las de Kraepelin (1901) y Roewer (1934) senalan que Ammotrechella geniculata aparte de las localidades que he comprobado de Colombia y Venezuela, se encontraria tambien en Cura?ao, Trinidad, Bahamas, Guadeloupe, Saint Vincent, Ecuador y Brasil. Seria necesa- rio con firmar si todas estas citas corresponden en realidad a A. geniculata. Por lo pronto, el material que he podido estudiar de Bahamas (machos y hembras) parece corresponder a una especie distinta. Respecto a Gluvia martha Karsch 1879, establezco en este trabajo la sinonimia con Ammotrechella geniculata, ya que he tenido la suerte de estudiar material tipico de ambas especies. Ni Kraepelin (1901), que tambien lo estudio, ni Roewer (1934) se decidieron en este sentido, a pesar de hacer notar ambos las simili- tudes entre dichos ejemplares. Especimenes estudiados.- VENEZUELA: “Orinoco” (Moritz leg.), syntypus hembra (ZMB 182); “Venezuela”, 20 de enero de 1903 (Briining), dos hembras (ZMH); San Jose de Avila (C. Vogl, un macho holotypus de Ammotrechula vogli Roewer 1952 (SMF 486); Estado Zulia, Maracaibo, 14 de abril de 1892 (E. Lep), una hembra (ZMH); Estado Miranda: Guatire, 1972 (M. A. Gonzalez Sponga), un macho y una hembra (MAGS 29); Estado Lara : Duaca, Fundo San Rafael, 26 de Febrero de 1968 (F. Delascio), una hembra (MHNLS 87 8); Distrito Federal, Caracas (sin fecha ni colector), dos hem- bras (MNHN), 1971 (M. A. Gonzalez Sponga), dos machos (MAGS 31); El Valle (V. Berthieu), una hembra (MNRJ 1347), Bellomonte, 26 de diciembre de 1978 (A. Rodriguez), un macho (MAGS 180), San Jose, 3 de agosto de 1959 (C. Ruiz Poleo) una hembra (MHNLS 573), La Pastora, 6 de diciembre de 1966 (P. Ojeda), una hembra (MHNLS 869), El Paraiso, 22 de diciembre de 1965 (W. Perez), un macho (MHNLS 578), 1 de abril de 1966 (W. Perez), un macho (MHNLS 579), diciembre de 1965 (Y. Ramirez), un juvenil (MHNLS 868), Altagracia, 20 de junio de 1966 (Y. Ramirez), una hembra (MHNLS 393), 5 de junio de 1966 (Y. Ramirez), un juvenil, (MHNLS 394), 30 de septiembre de 1966 (Y. Ramirez), una hembra (MHNLS 403), 28 de octubre de 1966 (Y. Ramirez), un macho (MHNLS 404), 26 de octubre de 1966 (Y. Ramirez), una hembra (MHNLS 405), 27 de agosto de 1964 (Y. Ramirez), un macho (MHNLS 586), 21 de marzo de 1966 (Y. Ramirez), una hembra (MHNLS 569), 29 de diciembre de 1965 (Y. Ramirez), una hembra (MHNLS 570), 13 de julio de 1964 (Y. Ramirez), una hembra (MHNLS 571), julio de 1964 (Y. Ramirez), una hembra (MHNLS 572), 12 de enero de enero de 1964 (Y. Ramirez), un macho (MHNLS 574), 22 de junio de 1964 (Y. Ramirez), una hembra (MHNLS 575), 23 de septiembre de 1964 (Y. Ramirez), una hembra (MHNLS 576), 8 de octubre de 1965 (Y. Ramirez), una hembra (MHNLS 577), 29 de noviembre de 1967 (Y. Ramirez), una hembra 134 THE JOURNAL OF ARACHNOLOGY (MHNLS 870), 18 de febrero de 1967 (G. Ramirez), una hembra (MHNLS 871), 18 de junio de 1967 (G. Ramirez), una hembra (MHNLS 872), 21 de marzo de 1967 (P. Ojeda), una hembra (MHNLS 873), 21 de marzo de 1967 (P. Ojeda), una hembra (MHNLS 874), 11 de mayo de 1967 (P. Ojeda), una hembra (MHNLS 875), 27 diciembre de 1967 (F. Yoris), un macho (MHNLS 876), 14 de noviembre de 1968 (M. Lentino), un macho (MHNLS 877), agosto de 1968 (L. Joly), un macho (MHNLS 879), 11 de febrero de 1968 (A. Perez), un macho (MHNLS 880), 3 de marzo de 1968 (Y. Ramirez), un macho (MHNLS 881), 28 de agosto de 1969 (W. Nazaret), una hembra (MHNLS 882), 13 de mayo de 1971 (M. Madriz), un macho (MNHLS 883). COLOMBIA: Santa Martha (Tetens leg.), una hembra holotypus de Gluvia martha Karsch 1879 (ZMB 2567); Sabanilla, 30 de agosto de 1908 (Moll leg.), un juvenil (ZMH); Cartagena (sin fecha ni colector), una hembra (MNHN). Xenotrecha, genero nuevo Cleobis Kraepelin 1899: 239 (no Oeobis Dana 1847). Ammotrecha Kraepelin 1901: 113 (no Ammotrecha Banks 1900). Ammotrechella Roewer 1934: 595 (en parte). Especie tipo: — Xenotrecha huebneri (Kraepelin 1899) por monotipia). Etimologia.— El nombre generico Xenotrecha proviene de la conjuncion del prefijo griego Xenos (extrano, como referencia al particular flagelo del macho) y trecha (trechos: correr, de donde deriva Ammotrechidae). Distribucion.-Venezuela. Diagnosis.— Ammotrechidae con los tarsos II y III con dos segmentos y el tarso IV con tres segmentos. Espinulacion de los tarsos II y III: 2.2.2/ 1 6 2.2. 2/ 1.1; espinulacion del tarso IV: 2.2. 2/2/0. Dedo movil de los queliceros con diente basal interno. Macho con ctenidios en el I y II esternitos espiraculares. En ambos sexos, dos dientes anteriores en el dedo fijo de los queliceros. Flagelo del macho con un pelo plumoso en la cara interna. Xenotrecha huebneri (Kraepelin 1899), nueva combination (Figs. 18-28) Cleobis hubneri Kraepelin 1899b: 239;Weidner 1959: 109. Ammotrecha hubneri: Kraepelin 1901: 113. Ammotrechella hiibneri: Roewer 1934: 595; Goetsch und Lawtsch 1944: 80; Caporiacco 1951: 35. Ammotrechella hubneri: Muma and Nazario 1971: 507; Muma 1976: 25. Description del holotypus hembra.— Este ejemplar, aparentemente subadulto, se en- cuentra en regular estado de conservation: el abdomen esta roto longitudinalmente y falta el pedipalpo derecho. Medidas en milimetros: Tabla I. Coloration: ejemplar descolorido por la mala fijacion. Morfologia: Prosoma: propeltidio algo mas ancho que largo (indice largo/ancho: 0,87). Lobulos laterales prominentes, separados del propeltidio por un surco dorsal. Tuberculo ocular con los ojos separados algo menos de un diametro ocular. El ejemplar ha perdido todas las setas del prosoma. Queliceros (Figs. 18-19): dedo movil muy robusto, con el mucron corto y de base ancha, no hay curvatura dorsal. Denticion: diente anterior ligeramente mas chico que el principal, el intermedio pequeno y el basal interno apenas esbozado. Dedo fijo: borde dorsal con una marcada prominencia a la altura del cuarto diente basal externo. Mucron corto y de base ancha. Denticion: dos dientes anteriores de tamano similar, un diente intermedio pequeno, un diente principal de tamano similar a los anteriores, cuatro basales externos que en orden decreciente de tamano se ordenan 3- 1-4-2 y cuatro basales internos, 1 y 3 largos y aguzados, el 2 vestigial y el 4 soldado por la base al 3. Pedipalpos: protarso con 1.1. 1.1. 1.2.2 espinas, tibia sin MAURY— SOLIFUGOS DE COLOMBIA Y VENEZUELA 135 Figs. 18-20 -Xenotrecha huebneri (Kraepelin), holotypus hembra: 18, quelicero derecho, vista externa; 19, quelicero derecho, vista interna; 20, operculo genital. Figs. 21-26 .-Xenotrecha huebneri (Kraepelin), macho: 21, quelicero derecho, vista externa; 22, quelicero derecho, vista interna; 23, operculo genital y esternitos espiraculares I y II (semiesque- matico); 24, pedipalpo derecho, vista ventral; 25, tarso III derecho, vista ventral; 26, tarso IV izquierdo, vista lateral. 136 THE JOURNAL OF ARACHNOLOGY espinas. Patas: protarsos II y III con 1.1 espinas dorsoposteriores y 1.2 ventrales; protarso IV con 1.1.2 espinas ventrales. Tarsos II y III con dos segmentos y con la espinulacion: 2. 2. 2/ 1.1; tarso IV con tres segmentos y con la espinulacion: 2. 2. 2/2/0. Descripcion de un macho adulto (Parque H. Pittier, Venezuela).-Medidas en mili- metros: Tabla I. Coloracion: color castano oscuro, mas pronunciado en prosoma, tergitos, pedipalpos y faz dorsal de las patas; en los queliceros se notan cuatro lineas longitudinales oscuras. Morfologia: Prosoma: propeltidio levemente mas ancho que largo (mdice largo/ ancho: 0,96). Lobulos laterales separados del prosoma por un surco dorsal. Cupula ocular prominente, con ojos separados algo menos de un diametro ocular. Todo el propeltidio, los restantes segmentos del prosoma y los tergitos, con diminutas espiculas. Esternitos (Fig. 23): esternitos espiraculares I y II con sendas areas de ctenidios. Queliceros (Figs. 21-22, 27-28): caras lateral y dorsal con espiculas y con algunas fuertes setas en distal. Dedo movil muy robusto, mucron de base ancha, aguzado y muy curvado; no hay curva- tura dorsal. Denticion: diente anterior y diente principal de igual tamano, el intermedio pequeno, el basal interno bien destacado. Dedo fijo de borde dorsal fuertemente curvado, sin prominencia. Mucron de base ancha, muy aguzado. Denticion: dos dientes anteriores de tamano similar, un intermedio muy pequeno, un principal de tamano similar a los anteriores, cuatro basales externos que en orden decreciente de tamano se ordenan 3- 1-4-2- y cuatro basales internos, 1 y 3 bien desarrollados, 2 muy pequeno, 4 unido por la base al 3. El flagelo sobrepasa levemente el borde dorsal del dedo fijo, es alargado y estrecho y se extiende desde el tercio distal del mucron hasta la altura del primer diente basal interno. El anillo de fijacion no se ve tan bien por transparency como sucede en otros Ammotrechidae, pero el flagelo esta soldado al quelicero de una manera similar (Fig. 28). En la cara interna del flagelo, aproximadamente desde el centro, nace un pelo plumoso que se extiende hacia adelante hasta casi el extremo del flagelo (Fig. 27). Este pelo es identico a otros que en numero variable se encuentran en la cara interna del quelicero. Pedipalpos (Fig. 24): protarso y tibia cubiertos de diminutas espiculas y con espinas ventrales. En el protarso parece haber 10 u 1 1 pares de espinas, pero algunas de la serie interna faltan, quedando incompleta (en el protarso izquierdo, por ejemplo, hay 2. 2. 2. 2. 2. 1.1. 2. 2. 2. 2 espinas). Tibia con 5 pares de espinas, aqui tambien algunas de la serie interna faltan y hay otras muy pequenas, supernumerarias. Patas: protarsos II y III con 1.1.1 espinas dorsoposteriores, la proximal muy pequena, y 1.2 espinas ventrales. Tarsos (Figs. 25-26): II y III con la siguiente espinulacion: 2.2.2/ 1 y tarso IV: 2. 2. 2/2/0. Descripcion de un ejemplar hembra (Pardillar, Venezuela).— Coloracion: similar a la del ejemplar macho descrito precedentemente. Morfologia: propeltidio mas ancho que largo (mdice largo/ancho: 0,83). Prosoma y queliceros con las espiculas y setas mucho mas debiles que en el macho. Queliceros: dedo movil similar al del macho. Dedo fijo con una prominencia dorsal a la altura del tercer diente basal externo. Denticion similar a la del macho. Pedipalpo: protarso con escasas espinas en la serie interna (protarso izquierdo con 1.1. 1.1. 1.1. 1.1. 2 espinas). Tibia sin espinas. Espinulacion de los tarsos similar a la del macho. Variabilidad.— Solo he podido estudiar tres especimenes de Xenotrecha huebneri y la unica variacion de importancia que he notado es la correspondiente a la espinulacion de los tarsos II y III, que mientras que en le holotypus es: 2. 2. 2/1.1 en los otros dos ejemplares es: 2. 2. 2/1. La presencia de una espina “suplementaria” no tiene a mi parecer mayor importancia, y es posible que cuando se estudien mas especimenes esta varicion se encuentre en una proportion variable, tal como sucede enAmmotrechellageniculata. La espinulacion de los protarsos de los pedipalpos presenta tambien cierta variacion individ- ual, pero la que Muma and Nazario (1971) indican para esta especie es indudablemente un error. MAURY- SOLIFUGOS DE COLOMBIA Y VENEZUELA 137 Figs. 21 -28. -Xenotrecha huebneri (Kraepelin), macho: 27, quelfcero derecho, vista interna (de- talle); 28, quelfcero derecho, vista dorsal (detalle). Figs. 29-34 -Eutrecha longirostris, especie nueva, holotypus macho: 29, quelfcero derecho, vista externa; 30, quelfcero derecho, vista interna; 31, operculo genital y esternitos espiraculares I y II (semiesquematico); 32, pedipalpo izquierdo, vista ventral; 33, tarso II derecho, vista ventral; 34, tarso IV derecho, vista lateral. 138 THE JOURNAL OF ARACHNOLOGY Especimenes estudiados.— VENEZUELA: “Siid-Venezuela”, 25 de noviembre de 1898 (G. Hiibner- O. Schneider), una hembra subadulta holotypus (ZMH); Estado Aragua, Parque “Henry Pittier”, 13 de agosto de 1978 (A. R. de Gonzalez), un macho (MAGS 211), Pardillar, 15 de agosto de 1978 (J. A. Gonzalez), una hembra (MAGS 42). Eutrecha , genero nuevo Especie tipo \ -Eutrecha longirostris, e specie nueva por designation original. Etimologia.— El nombre generico Eutrecha proviene de la conjuncion del prefijo griego Eu : bello (como referencia al elegante aspecto de este solifugo) y trecha (trechos: correr, de donde deriva Ammotrechidae). Distribution.— Venezuela. Diagnosis.— Ammotrechidae con los tarsos II y III con dos segmentos y el tarso IV con tres segmentos. Espinulacion de los tarsos II y III: 2.2.2/1 6 2. 2. 2/1.1; espinulacion del tarso IV: 2.2.2/2/0. Dedo movil de los queliceros con diente basal interno. Macho con ctenidios en el esternito espiracular I. Macho sin dientes anteriores en el dedo fijo de los queliceros. Eutrecha longirostris, especie nueva (Figs. 29-34) Description del holotypus macho.— Medidas en milimetros: Tabla I. Coloration: color general castano claro, algo mas oscuro en prosoma y segmentos distales de los pedipalpos; patas y tergitos con leve esfumado mas oscuro; queliceros con cuatro lineas longitudinales mas oscuras. Morfologia: Prosoma: propeltidio tan ancho como largo (indice largo/ ancho: 1). Lobulos laterales poco prominentes, separados del propeltidio por un surco dorsal. Cupula ocular con escotadura anterior; ojos separados poco menos de un diametro ocular. Todo el propeltidio ornado de diminutas espiculas, algunas mas grandes se destacan en los hordes laterales y posterior. Los otros segmentos del prosoma y los tergitos tambien ornados con espiculas. Esternitos (Fig. 31): esternito espiracular I con dos areas de ctenidios. Queliceros (Figs. 29-30): caras laterales y dorsal con espiculas y con algunas largas setas en distal. Dedo movil bastante mas robusto que el dedo fijo. Mucron corto, sin curvatura dorsal. Dentition: un diente anterior pequeno, de base ancha, un intermedio mas chico, un principal algo mayor que el anterior y un basal interno. Dedo fijo de borde dorsal ligeramente sinuoso; mucron casi recto, con el extreme curvado hacia ventral. Dentition: faltan los dientes anteriores y el intermedio, el diente principal es pequeno, hay cuatro basales externos que en orden decreciente de tamano se ordenan: 3- 1-4-2 y cuatro basales internos, 1 y 3 largos y puntiagudos, 2 y 4 mas pequenos. El flagelo es alargado y estrecho, sobresaliendo en toda su longitud por encima del borde dorsal del dedo fijo, se extiende desde el cuarto distal del mucron hasta la altura del cuarto diente basal interno, el extremo distal se abre en forma de embudo. Anillo de fijacion eliptico, situado a la altura de los dientes basales internos 1 y 2. Pedipalpos (Fig. 32): protarso con 5 pares de espinas pero algunas pueden faltar (protarso derecho, por ejemplo, con 1.2. 1.2.2 espinas); tibia con 4 pares de espinas, irregularmente dispuestas. Patas: protarsos II y III con 1.1 espinas dorsoposteriores y 1.2 ventrales; protarso IV con 1.1.2 espinas ventrales. Tarsos (Figs. 33-34): tarsos II y III con dos segmentos y con la espinulacion: 2.2.2/ 1 6 2.2.2/ 1.1; tarso IV con tres segmentos y con la espinulacion: 2.2.2/2/0. MAURY- SOLIFUGOS DE COLOMBIA Y VENEZUELA 139 Comentarios.— Aunque he podido ver solamente un ejemplar de Eutrecha longirostris, el estudio de la espinulacion tarsal de este especimen corrobora mis apreciaciones sobre la variabilidad de este caracter. Mientras el tarso II derecho posee 2. 2. 2/ 1.1 espinas, en el tarso II izquierdo hay 2. 2.2/1; el tarso III derecho tiene 2. 2. 2/1 espinas y el tarso III izquierdo 2. 2. 2/ 1.1. Especimenes estudiados.- VENEZUELA: Departamento Vargas, D.F., Punta de Tarma, 6 de diciembre de 1978 (M. von Dagel), un macho holotypus (MAGS 167). Mummuciona Roe we r Mummuciona Roewer 1934: 590; Mello-Leitao 1938a: 17; Caporiacco 1951: 34; Lawrence 1954: 122; Muma 1976: 24. Especie tipo —Mummuciona simoni Roewer 1934, por designation original. Distribution.— Venezuela, Colombia, (Peru ? ). Diagnosis.— Ammotrechidae con los tarsos II y III con dos segmentos y el tarso IV con tres segmentos. Espinulacion de los tarsos II y III: 2. 2. 2/2.1 6 2. 2. 2/1.1; espinulacion del tarso IV: 2. 2. 2/2/2 6 2.2.2/2/1. Dedo movil de los queliceros sin diente basal interno. Dedo fijo de los queliceros con el segundo diente basal interno austente o vestigial; en ambos sexos hay dos dientes anteriores. Macho con 'ctenidios en los esternitos espira- culares I y II. Mummuciona simoni Roewer 1934 (Figs. 35-43) Cleobis limbata: Kraepelin 1899a: 378 (en parte, no Cleobis limbata Lucas 1835). Mummuciona simoni Roewer 1934: 590; Goetsch und Lawatsch 1944: 80: Caporiacco 1951: 35; Lawrence 1954: 122 (partim); Muma 1976: 24. ? Mummuciona marcuzzii Caporiacco 1951: 34. Description de un syntypus hembra.— Este ejemplar se encuentra en malas condiciones de conservation, el abdomen esta desprendido a nivel del tergito 3 y faltan varias de las patas. Medidas en milimetros: Tabla I. Coloration: ejemplar destenido por la mala fijacion. Morfologia: Prosoma: propeltidio algo mas ancho que largo (mdice largo/ancho: 0,90). Lobulos laterales separados del propeltidio por un surco dorsal. Cupula ocular con los ojos separados algo menos de un diametro ocular. El propeltidio ha perdido casi toda la cubertura pilosa, solo se ven algunas diminuats espiculas diseminadas. Queliceros (Figs. 35-36): dedo movil con el mucron sin curvatura dorsal. Denticion: un diente anterior, un diente intermedio pequeno y un diente principal de tamano similar al anterior; no hay diente basal interno. Dedo fijo con una prominencia a nivel del primer diente basal externo; mucron corto, poco curvado. Denticion: dos dientes anteriors de tamano similar, un intermedio pequeno, un principal de tamano ligeramente mayor que los anteriores, cuatro basales externos que en orden decreciente de tamano se ordenan 3- 1-4-2 y tres basales internos (falta el segundo). Pedipalpos: protarso con 8 6 9 pares de espinas; tibia con 6 pares de espinas, pero algunas pueden faltar (la tibia derecha, por ejemplo, tiene 1.1. 2. 2. 2. 2 espinas). Patas: protarsos II y III con 1.1.1 espinas dorsoposteriores y 1.1.2 espinas ventrales; protarso IV con 1.1.2 espinas ventrales. Tarsos II y III con dos seg- mentos y con la espinulacion: 2. 2. 2/ 1.1 y tarso IV con tres segmentos y con la espinu- lacion 2.2. 2/2/1 . 140 THE JOURNAL OF ARACHNOLOGY Descripcion de un ejemplar macho (Valledupar, Colombia) -Medidas en miimetros: Tabla I. Coloration: color general castano, mas acentuado en prosoma, tergitos, pedi- palpos y articulos distales de las patas. Morfologia: Prosoma: propeltidio un poco mas ancho que largo (mdice largo/ancho: 0,87). Lobulos laterales prominentes, separados del propeltidio por un surco dorsal. Tuberculo ocular con los ojos separados menos de un diametro ocular. Todo el propeltidio con fuertes setas espiniformes terminadas en furcula. Los otros segmentos del prosoma y los tergitos con el mismo tipo de setas. Esternitos (Fig. 40): con largos pelos sedosos, esternitos espiraculares I y II con algunos ctenidios aislados, en el I hay aproximadamente 10, en el II hay seis. Queliceros (Figs. 38-39): caras laterales y dorsal con el mismo tipo de setas terminadas en furcula del prosoma. Dedo movil con el mucron alargado, puntiagudo y sin curvature dorsal. Denticion: un diente anterior y un diente principal de tamano similar, el intermedio muy pequeno; no hay diente basal interno. Dedo fijo con el borde dorsal levemente convexo; mucron corto y dirigido hacia adelante. Denticion: dos dientes anteriores, de los cuales el primero mas pequeno y separado del segundo por una amplia diastema, un intermedio de tamano similar al primero anterior, un principal que es el mayor de la serie, cuatro basales externos que en orden decreciente de tamano se ordenan 1 -3-4-2 y tres basales internos (falta el segundo). Flagelo alargado, sobresaliendo en toda su extension al borde dorsal del dedo fijo; se extiende desde aproximadamente la base del mucron hasta la altura del tercer diente basal interno. Anillo de fijacion eliptico, situado a nivel del primer diente basal interno. El flagelo termina en una punta ligeramente bifucada, la punta dorsal algo mas larga que la ventral. Pedipalpos (Fig. 41): protarso con numerosas setas de diferente grosor, que enmascaran mucho las espinas, parece haber 8 pares de espinas, dificiles de distinguir; en la tibia parece haber 5 pares de espinas, aunque algunas faltan (tibia derecha con 2. 1.1. 1.1; tibia izquierda con 2. 2. 1.1. 2 espinas). Patas: protarsos con la misma espinulacion que en el holotypus. Tarsos (Figs. 42-43): II y III con la espinulacion: 2.2.2/2.1; tarso IV con la espinulacion: 2. 2. 2/2/2. Variabilidad.-En los cinco ejemplares que he podido estudiar d e Mummuciona simoni. he comprobado variaciones significativas en la espinulacion de los tarsos II, III y IV y tambien en la espinulacion de los pedipalpos. En cuatro ejemplares hay austencia absoluta del segundo diente basal interno del dedo fijo de los queliceros, pero en una hembra, y solamente en el quelicero izquierdo, hay un vestigio de dicho diente. Comen tarios.— El material tipico mencionado por Roewer (1934) consistia en dos hembras, depositadas en la coleccion de aracnidos del MNHN, pero en la actualidad se encuentra una sola de ellas en el frasco correspondiente. Caporiacco (1951) describe para Venezuela una segunda especie de Mummuciona, que denomina M. marcuzzii. Este autor cita material proveniente de las siguientes localidades: Distrito Federal: Tacagua, Isla Margarita: Prolamar, Lara: Barquisimeto y Sucre: Cumana. Todo este material, de- positado otrora en el Museo de Biologia, Universidad Central de Venezuela, Caracas, parece haber sido destruido o extraviado. Confrontando la descripcion original de Mum- muciona marcuzzi con el material que he estudiado de M. simoni, no veo diferencias substanciales para poder separar ambas especies. Al no tener por el momento otra evi- dencia, cito a M. marcuzzii con un interrogante en la sinonimia deM. simoni. Respecto a la cita de Lawrence (1954) del genero Mummuciona para el norte del Peru (Lobitos), es algo que debera confirmarse. El unico material estudiado consiste en una hembra en regulares condiciones de conservation, tanto que Lawrence no considero adecuado darle categoria especifica. MAURY- SOLIFUGOS DE COLOMBIA Y VENEZUELA 141 Figs. 35-37 -Mummuciona simoni Roewer, syntypus hembra: 35, quelicero derecho, vista externa; 36, quelicero derecho, vista interna; 37, operculo genital. Figs. 38-42 .—Mummuciona simoni Roewer, macho: 38, quelicero derecho, vista externa; 39, quelicero derecho, vista interna; 40, operculo genital y esternitos espiraculares I y II (semiesque- matico); 41, pedipalpo izquierdo, vista ventral; 42, tarso III izquierdo, vista ventral; 43, tarso IV izquierdo, vista lateral. Figs. 44-46. -Ammotrechula sp., hembra; 44, quelicero derecho, vista externa; 45, pedipalpo izuierdo, vista ventral; 46, operculo genital. 142 THE JOURNAL OF ARACHNOLOGY Especimenes estudiados.— VENEZUELA: “Orinoco” (sin fecha ni colector), una hembra syntypus (MNHN); “Venezuela”, 1899 (F. Geay), una hembra (MNHN). COLOMBIA: Departmento Cesar, Valledupar, 3 de septiembre de 1968 (B. Malkin), un macho (AMNH), 22-24 de mayo de 1968 (B. Malkin), una hembra (AMNH); Departmento Magdalena : Santa Marta, Rodadero, 27 de abril de 1968 (B. Malkin), una hembra (AMNH). Ammotrechula sp. (Figs. 44-46) Description de un ejemplar hembra -Longitud total: 18 mm. Coloration: color cas- tano claro, algo mas oscuro en el dorso. Queliceros (Fig. 44): dedo movil con el mucron sin curvatura dorsal. Dentition: un diente anterior, un intermedio pequeno y un principal, ligeramente mayor que el anterior; basal interno bien marcado. Dedo fijo con una mar- cada prominencia a nivel del primer diente basal externo. Dentition: dos dientes an- teriores de tamano similar, un intermedio, un principal algo mas grande que los anteriores, cuatro basales externos de tamano parejo y cuatro basales internos. Los queliceros son muy similares a los que presenta la hembra de Ammotrechella geniculata. Pedipalpos (Fig. 45): protarso con 10 u 11 pares de espinas, series interna incompleta y con algunas espinas supernumerarias entre las dos series principales; tibia con 8-9 pares de espinas, tambien aqui la serie interna es incompleta y hay espinas supernumerarias. Patas: pro- tarsos II y III con 1.1.1 espinas dorsoposteriores y 1.1.2 ventrales; protarso IV con 1.1.2 espinas ventrales. Tarsos II y III con dos segmentos y con la espinulacion: 1.2. 2/2.1; tarso IV con cuatro segmentos y con la espinulacion: 2.2/2/2/2. Operculo genital indicado en la figura 46. Comentarios.— La fait a de un especimen macho me impide por el momento ubicar mas concretamente a esta especie. Segun Muma (1976); el genero Ammotrechula tiene actual- mente unas 13 especies descriptas, la mayoria de America del Norte: sur de los Estados Unidos y Mexico. Para America del Sud se han citado dos especies: Ammotrechula vogli Roewer 1952, de Venezuela y que en este trabajo considero bajo la sinonimia de Ammotrechella geniculata, y Ammotrechula gervaisi. (Pocock 1899), del Ecuador (? )y Colombia. He podido estudiar el ejemplar tipo de esta ultima especie (una hembra) pero difiere del que acabo de describir en varios caracteres: protarsos de los pedipalpos con solamente 6 pares de espinas; tibia inermes y una configuration de los queliceros leve- mente distinta. No he podido encontrar en el BMNH los ejemplares mencionados por Pocock para Colombia, aunque este autor manifiesta que estaban “. . . too mutilated and shrivelled for accurate determination. . .”. Especimenes estudiados.-COLOMBIA: Rio Frio, 1910 (F.C.P.), una hembra (USNM). AGRADECIMIENTOS Este trabajo ha sido posible realizarlo gracias a la colaboracion de numerosos colegas, que facilitaron mi labor enviandome colecciones y material tipico. Por esta razon estoy profundamente reconocido a las siguientes personas: Dr. M. Moritz. Zoologisches Museum, Humboldt Universitat, Berlin (ZMB); Dra. G. Rack, Zoologisches Staatsinstitut und Zoologisches Museum, Hamburgo (ZMH); Dr. M. Grasshoff, Senckenberg Museum, Frankfurt (SMF); Dr. F. Wanless y Dr. P. Hillyard, British Museum (Natural History), Londres (BMNH); Dra. J. Heurtault, Museum National d’Histoire Naturelle, Paris (MNHN); Dr. N. Platnick, American Museum of Natural History, Nueva York (AMNH); MAURY- SO LIFUGOS DE COLOMBIA Y VENEZUELA 143 Dr. R. Crabill, National Museum of Natural History, Washington (USNM) y Dra. A. T. da Costa, Museu Nacional, Rio de Janeiro (MNRJ). Mi especial agradecimiento al Dr. M. A. Gonzalez Sponga, Institute Pedagogico, Caracas, por los tramites y el envio de su colec- cion particular (MAGS) y la del Museo de Historia Natural La Salle, Caracas (MHNLS); tambien quedo reconocido a la Dra. Alda Gonzalez (La Plata), que transporto gentilmente a Buenos Aires parte de esta coleccion. LITERATURA CITADA Butler, A. G. 1873. List of the species of Galeodides, with description of a new species in the collection of the British Museum. Trans. Ent. Soc. London: 415-425. Caporiacco, L. di. 1951. Studi sugli aracnidi del Venezuela. I Parte: Scorpiones, Opiliones, Solifuga y Chernetes. Acta Biol. Venezuelica, 9(1): 1-46. Goetsch, W. und E. Lawatsch. 1944. Beitrage zur biologie und verbreitung sudamerikanischer walzen- spinnen. Zool. Anz., 144(5-6):73-90. Karsch, F. 1879. Sieben neue Arachniden von St. Martha. Stettin. Ent. Zeitg., 40:106-109. Karsch, F. 1880. Zur Kenntniss der Galeodiden. Arch. Naturg., 46(l):228-243. Koch, C. L. 1842. Systematische uebersicht liber die familie der Galeoden. Arch. Naturg., 8(l):350-356. Koch, C. L. 1848. Die Arachniden, 15:1-136. Kraepelin, I. 1899a. Catalogue des Solifugues des collections du Museum d’Histoire Naturelle de Paris. Bull. Mus. Hist. Nat. Paris, 7:376-378. Kraepelin, K. 1899b. Zur systematik der Solifugen. Mitt. Naturhist. Mus. Hamburg, 16:195-259. Kraepelin, K. 1900. Ueber einige neue gliesderspinnen. Abh. Natur. Ver. Hamburg, 16(1): 1-17. Kraepelin, K. 1901. Palpigradi und Solifugae. Das Tierreich, 12:1-159. Lawrence, R. F. 1954. Some Solifugae in the collection of the British Museum (Natural History). Proc. Zool. Soc. London, 124(1): 1 11-124. Maury, E. A. 1976. Nuevos solifugos Ammotrechidae de la Argentina (Arachnida, Solifugae). Physis C, 35(90):87-104. Maury, E. A. 1977. Notas sobre la sistematica y distribution geografica de Procleobis patagonicus (Holmberg, 1876)(Solifugae, Ammotrechidae, Saronominae). Physis C, 36(92):283-293. Maury, E. A. 1980. Presencia de la familia Daesiidae en America del Sur con la description de un nuevo genero (Solifugae). J. Arachnol., 8(l):59-67. Mello-Leitao, C. 1938a. Solifugos de Argentina. An. Mus. Arg. Cienc. Nat., 50 (Ent. 156) : 1-32. Mello-Leitao, C. 1938b. Notas sobre solifugos argentinos. Not. Mus. La Plata, Zool., 3(15):265-271. Muma, M. H. 1951. The arachnid order Solpugida in the United States. Bull. Amer. Mus. Nat. Hist., 97(2):35-141. Muma, M. H. 1970. A synoptic review of North American, Central American, and West Indian Solpugida (Arthropoda: Arachnida). Arthopods of Florida, 5:1-62. Muma, M. H. 1976. A review of solpugid families with an annotated list of Western Hemisphere solpugids. WRI/SCI, Silver City, 2(1): 1-33. Muma, M. H. and M. L. Nazario. 1971. New solpugids (Arachnida: Solpugida) from Puerto Rico. J. Agric. Univ. Puerto Rico, 55(4):506-512. Pocock, R. I. 1895. Notes on some of the Solifugae contained in the collection of the British Museum, with description of new species. Ann. Mag. Nat. Hist., 6 ser., 16:74-98. Roewer, C. Fr. 1934. Solifugae, Palpigradi, in: “Bronns Klassen une Ordnungen des Tierreichs” 5, 4(4):l-723. Roewer, C. Fr. 1941. Solifugen 1934-1940. Veroff. Deutsch. Kol.-Uber.-Mus. Bremen, 3(2):97-192. Roewer, C. Fr. 1952. Neotropische Arachnida Arthrogastra zumeist aus Peru. Senckenbergiana, 33(l-3):37-58. Weidner, H. 1959. Die entomologischen sammlungen des Zoologischen Staatsinstituts und Zoolo- gischen Museum Hamburg, I. Teil. Pararthopoda und Chelicerata I. Mitt. Hamburg. Zool. Mus. Inst., 57:89-142. Zilch, A. 1946. Katalog der Solifugen (Arach.) des Senckenberg-Museums. Senckenbergiana, 27(4-6): 119-154. Manuscript received April 1981, revised May 1981. Capocasale, R. M. 1982. Las especies del genero Porrimosa Roewer, 1959 (Araneae, Hippasinae). J. J. Arachnol., 10:145-156. LAS ESPECIES DEL GENERO PORRIMOSA ROEWER, 1959 (ARANEAE, HIPPASINAE)1 Roberto M. Capocasale Institute de Investigaciones Biologicas Clemente Estable Division Zoologia Experimental Av. Italia 3318 Montevideo, Uruguay ABSTRACT The name Porrimosa Roewer is selected to replace Porrima Simon, which is preoccupied. Of the seven nominal species studied four are considered to be species inquirenda. This is because the holo- type of P. callipoda cannot now be found, and for the three others (P. diversa, P. glieschi and P. securifera ) the descriptions were based upon juveniles. Three species are fully described, including both sexes for the best known species, P. lagotis (Mello-Leitao), not the P. lagotis of Holmberg. For P. castanea only the female is known, and for P. harknessi only the male. Four new combinations are indicated, among which one species, P. securifera, is transferred from the genus Lycosa to Porrimosa. RESUMEN Porrimosa Roewer es el nombre sustituto de Porrima Simon, pues este ultimo es un nombre preocupado. De siete especies de Porrimosa estudiadas, cuatro se consideran species inquirenda porque el holotipo de Porrimosa callipoda no ha sido hallado hasta ahora y los otros tres holotipos conocidos ( Porrimosa diversa, Porrimosa glieschi y Porrimosa securifera ) se describieron de formas inmaduras. Se describen tres especies, de las cuales, se conocen ambos sexos, en solo una de ellas: P. lagotis (Mello- Leitao). En las otras dos solo se conoce la hembra en P. castanea, y el macho en P. harknessi. Se indican cuatro nuevas combinaciones, entre las cuales una especie (. Lycosa securifera Tullgren) se transfiere del genero Lycosa al genero Porrimosa. INTRODUCCION Porrima es un genero creado por Simon en 1898 cuyas especies hasta ahora fueron halladas exclusivamente en America del Sur; su especie tipo es Podophthalma diversa 0. P.-Cambridge, 1877. La diagnosis que Simon (1898) dio para Porrima fue la siguiente: “Oculi quatour antici in lineam latam, valde procurvam, medii a lateralibus quam inter se remotiores, laterales mediis majores prominuli et utrinque prope marginem clypei siti. Oculorum linea postica antica paulo brevior. Chelarum margo inferior quadridentatus. Partes oris, pedes mamillaeque Hippasae ”. ^rabajo realizado con la ayuda del “Programa Regional de Desarrollo Cientifico y Tecnologio” de la Organization de los Estados Americanos (O. E. A.). 146 THE JOURNAL OF ARACHNOLOGY Actualmente el estudio de este genero presenta problemas formales, de nomenclatura, y facticos, de observation, tan estrechamente ligados que no es posible estudiarlos separa- damente. Segun Neave (1940) el nombre Porrima fue usado para un lepidoptero, 21 anos antes que lo hiciera Simon. De acuerdo con el ICZN (Art. 53) Porrima debe ser susti- tuido. Sin embargo, reiteradamente Mello-Leitao (1941, 1942, 1943, 1945, 1947, 1948, 1949) uso el nombre Porrima. Roewer (1954) creo para Porrima callipoda Mello-Leitao un genero nuevo , Porrimula, usando como base la description original que establecio: “Quelicerios con quatro dentes fortes”. En 1959, el mismo autor creo otro genero nuevo , Porrimosa, para las especies que tienen 3 dientes en el borde posterior interno de sus queliceros, y las distancias inte- roculares de los ojos medianos posteriores iguales a un diametro de dichos ojos. En sus revisiones Roewer (1954b, 1959), puso enfasis en la importancia de los dientes en el borde posterior interno de los queliceros y en las medidas del area ocular, para caracte- rizar sus diagnosis genericas. En 65 especimenes estudiados de Porrimosa , de varias especies y areas geograficas, inclufdos 5 tipos, -se hallo que el 97.6% de los casos tuvieron 3 dientes en el borde posterior interno de sus queliceros. Este caracter, por lo tanto es fundamental para definir el genero. No ocurre lo mismo con las relaciones del area ocular. En la poblacion antes men- cionada, se hallo que la relation de las distancias del area ocular no se cumple en el 92.36% de los casos. De donde se concluye que estas relaciones no son un caracter que sirva para definir a Porrimosa. Pero, ademas, segun lo observado, fue un error considerar que ‘ Porrima’ ’ tiene 4 dientes en el borde posterior interno de sus queliceros. En efecto, ni los sintipos de Podophthalma diversa 0. P.-Cambridge ni los especimenes de Porrima diversa estudiados por Simon, apoyado en los cuales confecciono la diagnosis antes anotada, tienen 4 dientes posteriores quelicerales, sino 3. Todo esto hace que, de hecho, las divisiones genericas de Roewer no sean validas pues estan apoyadas en caracteres estadfsticamente insignificantes. Pero, formalmente, sus nombres deben ser utilizados. Porrimula Roewer deberia sustituir a Porrima, pero existe un inconveniente: el holo- tipo de P. callipoda no ha podido ser hallado -seguramente este perdido-; por esta causa Porrimula debe ser considerado como nomen dubium. El nombre inmediato a usar es Porrimosa, que efectivamente posee condiciones formales y facticas como para ser susti- tuto de Porrima. ESPECIES DE UBICACION DUDOSA EN EL GENERO PORRIMOSA Mello-Leitao (1947) argumento que “As especies sul americanas atribuidas ao genero Tetragonophthalma pertenecem realmente ao genero Porrima ”. Si se respeta ese juicio se deben incluir en el genero las siguientes especies: Tetragonophthalma freiburgensis Key- serling, 1877, Tetragonophthalma obscura Keyserling, 1891 y Tetragonophthalma spinipes Taczanowski, 1873. En la literatura no se hallaron resultados explicando el fundamento de esa conclusion. Los resultados de la observation del material determinado por Mello-Leitao hallado en colecciones (los cuales no avalan la hipotesis de dicho autor) son los siguientes: Porrima freiburgensis: Mello-Leitao, 1943: 164. Dos especimenes de Porto Alegre, Brasil. (1 hembra inmadura, 1 inmaduro) examinados, depositados en el Museu Nacional de Rio de Janeiro. CAPOCASALE- ESPECIES DEL GENERO PORRIMOSA 147 Porrima granadensis: Mello-Leitao, 1941: 278. Dos especimenes de La Uvita. Colombia (1 macho; 1 hembra) examinados, identificacion erronea, depositados en el Museu Nacional de Rio de Janeiro. Porrima ohscura: Mello-Leitao, 1943: 165. Cinco especimenes de: Itapicuru (1 hembra) identificacion erronea; Campino Grande (3 hembras inmaduras; 1 inmaduro) Exami- nados, Brasil, depositados en el Museu Nacional de Rio de Janeiro. Hasta tanto los “tipos” de dichas especies sean localizados y observados, en este estudio se excluyen del genero. 5 Figs. 1-5. 1, 2: Porrimosa castanea (Mello-Leitao) Holotipo hembra (sin numero; MNRJ); 1, espermatecas; 2, epigino. 3-5: Porrimosa lagotis (Mello-Leitao); 3, espermatecas (N° 14945; MLP) Lectotipo; 4, epigino (Curitiba, Brasil); 5, epigino (Maldonado, Uruguay), b. c. = bolsa copulatoria, e. = espermatecas, g. m. = guia mediana, g. t. = guia transversa. 148 THE JOURNAL OF ARACHNOLOGY Porrimosa Roewer, 1959 Porrima Simon 1898: 327. Porrimula Roewer 1954: 313. Porrimosa Roewer 1959: 1001, 1002. Ojos: primera fila procurva, los antero-laterales sobre promontorios, su diametro es mayor que el de los antero-medianos; segunda fila de diametro semejante al de los antero- laterales; tercera fila apoyados sobre promontorios pero dirigidos hacia atras, su diametro es semejante al de los medianos (Fig. 14). Queliceros: destacados, 3 dientes anteriores y 3 posteriores en cada uno (Fig. 13). Labio: mas largo que ancho. Patas: par IV el mas largo de todos. Hileras: biarticuladas, el artejo terminal espatulado. Epigino: en forma de T invertida, el eje longitudinal de la guia mediana 3-4 veces mas corto que el de la guia transversa (Fig. 2-5). Espermatecas: con numerosos tuberculos apicales, arqueadas en angulo mas o menos recto, los vertices de dichos angulos dirigidos hacia el centro (Fig. 1-3). Palpo: apofisis mediana breve, apofisis lateral del conductor curvada, terminada en una punta muy aguda (Fig. 6-10). CLAVE DE ESPECIES HEMBRAS 1. Epigino con la guia transversa dos veces mas ancha que largo (Fig. 2). Espermatecas con su bolsa copulatoria bilobulada (Fig. 1). Dorso del abdomen castano, salpicado con puntos amarillos (machos, desconocidos) Porrimosa castanea Epigino con la guia transversa cuatro veces mas ancha que larga (Fig. 4-5). Esper- matecas con su bolsa copulatoria simple (Fig. 3). Dorso del abdomen con una franja mediana (Fig. 12). Centro de Argentina, sur del Brasil y Uruguay • Porrimosa lagotis MACHOS 1. Palpo con el cymbium y el conductor largo y fino. Dorso del abdomen castano rojizo; ventral castano oscuro con diseno que sigue el modelo de la Fig. 12 (hembras desco- nocidas). Colombia y sur del Peru Porrimosa harknessi Palpo con el cymbium mas ancho que largo, conductor casi tan corto como ancho. Dorso del abdomen como el diseno de la Fig. 1 1 ; ventral amarillo claro Porrimosa lagotis Porrimosa callipoda (Mello-Leitao, 1934), nueva combination Porrima callipoda Mello-Leitao 1934: 405, Fig. 5-6; Bonnet 1958: 3765. Porrimula callipoda: Roewer 1954: 313; 1959: 1005; 1961: 16. Tetragonophthalma callipoda: Biicherl y Lucas 1972: 267. Comentarios.— Holotipo presumiblemente perdido. Segun Mello-Leitao, una hembra de Riberao Claro, Matto Grosso, Brasil, depositado en el Instituto Butantan. CAPOCASALE— ESPECIES DEL GENERO PORRIMOSA 149 Los especimenes Nos. 1002 6 1003 (examinados) de la coleccion del Instituto Butantan se estima que no son, alguno, el tipo de Porrima callipoda como lo dijeron Biicherl y Lucas (1972). Se considera species inquirenda. Porrimosa castanea (Mello-Leitao, 1942) Fig. 1,2, 12, 14, Map. 1 Porrima castanea Mello-Leitao 1942: 432, Fig. 6. Porrimosa castanea: Roewer 1959: 1002, 1005; 1961: 16. Diagnosis.— E specie facilmente separable de las restantes del genero por la coloracion general del abdomen. La hembra se aleja de Porrimosa lagotis por tener: a) epigino mas grande, fundamentalmente el eje longitudinal de la guia transversa es mayor, b) bolsa copulatoria doble y c) espermatecas morfologicamente diferentes (Fig. 1-2). Hembra.— Largo total: 17.7 mm. Cefalotorax: largo 7.5 mm, ancho 5.5 mm, bordes laterales castano, mancha submarginal continua amarillo claro; 6 lineas radiales castano convergentes hacia el surco toracico. Esternon: amarillo. Queliceros: castano naranja. Patas: amarillo, metatarsos y tarsos de pata I sin escopula. Abdomen: dorsal castano salpicado de puntos amarillos, no se observa diseno; areas laterales, el mismo tinte que el dorso pero los puntos amarillos mas separados, lo cual las hace mas claras; ventral, una franja mediana castano bordeada de puntos amarillos (Fig. 12). (Ver comentarios). Epigino y espermatecas: como los esquemas de las figuras 1 y 2. Comentarios.-Holotipo hembra de La Merced, Peru, examinado, depositado en el Museu Nacional de Rio de Janeiro. La descripcion anterior fue hecha utilizando el unico especimen hallado, de esta especie, determinado y etiquetado como “typus” por Mello-Leitao. Existen diferencias fundamentals en la coloracion, si se compara la descripcion de Mello-Leitao, atribuibles seguramente al tiempo que el especimen permanecio en el conservador. Porrimosa diver sa (O. P. -Cambridge, 1877), nueva combinacion Podophthalma diversa P.-Cambridge 1877: 572, Pla. 57, Fig. 9a-b; Simon 1898: 327, Fig. 332-333. Tetragonophthalma diversa: Keyserling 1891: 255. Porrima diversa: Roewer 1954: 313; Bonnet 1958: 3765; Mello-Leitao 1941: 201; 1942: 383; 1943: 164; 1945: 221; 1947: 266 (identificacion erronea); 1948: 153 (identificacion erronea); 1949: 4; Brady 1962: 129, Fig. 11. Comentarios.— Cinco sintipos de Minas Gerais, Brasil (5 inmaduros) examinados, depositados en Hope Department Entomology. University Museum. Muchos especimenes que fueron indicados para otras localidades de Brasil, no men- cionados en “Material examinado”, no han sido hallados en el Museu Nacional de Rio de Janeiro, por lo que se presume que estan perdidos. Se considera species inquirenda . Material examinado.— Treinta y un especimenes, determinados por Simon, de : La Pura, Santarem, Cameta, Brasil, (3 hembras, 2 hembras inmaduras) identificacion erronea; Loja, Ganjou, Brasil (1 hembra) identificacion erronea. America Meridional (sic) (8 hembras inmaduras, 5 machos inmaduros, 9 inmaduros); Matto Grosso, Brasil (1 hembra inmadura); Venezuela (2 hembras) identificacion erronea, depositados en el Museum National d’Histoire Naturelle, Paris. Dos especimenes, deter- minados por Mello-Leitao, de Arocena, Caraguatay, Santa Fe, Argentina (1 hembra inmadura); de- 150 THE JOURNAL OF ARACHNOLOGY positados en el Museo de La Plata. Un especimen, determinado por Mello-Leitao, de Basail, Chaco, Argentina (juvenil), depositado en el Museo de La Plata. Dos especimenes, determinados por Mello- Leitao, de Manantiales, Corrientes, Argentina (inmaduros), depositados en el Museo de La Plata. Tres especimenes, determinados por Mello-Leitao, de Curitiba, Brasil (1 hembra inmadura), depositado en el British Museum (Natural History) y (2 hembras) identificacion erronea, depositados en el Museu Nacional de Rio de Janeiro. Un especimen, determinado por Mello-Leitao, de Canister Falls, Guyana (hembra) identificacion erronea, depositado en el British Museum (Natural History). Porrimosa glieschi (Mello-Leitao, 1926) Porrima glieschi Mello-Leitao 1926: 2; Roewer 1954: 313; Bonnet 1958: 3765. Porrimosa glieschi: Roewer 1959: 1005; 1961: 16. Comentarios.— Holotipo de Rio Grande do Sul, Brasil (1 hembra inmadura) exami- nado, depositado en el Museu Nacional de Rio de Janeiro. Se considera species inquirenda. Mapa 1 .— Distribucion geografica, conocida, del genero Porrimosa . El circulo bianco ( Porrimosa sp.) indica localidades marcadas sobre la base de especimenes inmaduros. El signo de interrogacion(? ) senala indicacion imprecisa. C APOCAS ALE— ESPECIES DEL GENERO PORRIMOSA 151 Porrimosa harknessi (Chamberlin, 1916) Fig. 9-10, Map. 1 Porrima harknessi Chamberlin 1916: 280, Tab. 23, Fig. 2-6; Roewer 1954: 313; Bonnet 1958: 3765; Brady 1962: 129, Fig. 33. Porrimosa harknessi: Roewer 1959: 1001, 1005; 1961: 16. Porrima granadensis: Mello-Leitao, 1941: 278. Diagnosis.— Los machos de esta especie se diferencian de Porrimosa lagotis por la coloration general, mas rojiza en P. harknessi. El abdomen, ventralmente, es castano oscuro y con diseno que sigue el modelo de la figura 12, mientras que en P. lagotis es simplemente amarillo claro. Los palpos tambien son diferentes si se comparan sus dimen- siones. El cymbium y el conductor de P. harknessi son mas largos y finos que en P. lagotis. Macho.— Largo total: 15.0 -15.5 mm. Cefalotorax: largo 7.0 -7.3 mm, ancho 5.0 -5.5 mm (2 especimenes medidos); hordes laterales castano oscuro; mancha submarginal continua amarillo; 6 lfneas radiales castano convergentes hacia el centro del surco toracico, las 2 posteriores mas anchas. Esternon: castano naranja con una franja mediana negro. Queliceros: castano naranja. Patas: castano naranja, metatarsos y tarsos castano naranja, metatarso y tarso de pata I escopulados. Abdomen: una franja mediana castano oscuro, diseno muy semejante al de figura 1 1 ; areas laterales castano muy oscuro; ventral una franja mediana castano bordeada de puntos irregulares amarillo (el diseno sigue el modelo de la figura 12). Palpos como los esquemas de las figuras 9-10. Comentarios.— Holotipo macho y 2 paratipos (1 macho; 1 hembra inmadura) de Haudquina, Peru, examinados, depositados en el Museum of Comparative Zoology, Har- vard University. Material examinado.-Dos especimenes de La Uvita, Colombia (1 macho; 1 hembra) examinados, depositados en el Museu Nacional de Rio de Janeiro, determinados por Mello-Leitao como Porrima granadensis (la hembra esta muy deteriorada; como no se tiene certeza de su coespecificidad no se uso para describir). Porrimosa lagotis (Mello-Leitao, 1941), nueva combination Fig. 3-5, 6-8, 11, Map. 1 Porrima lagotis Mello-Leitao. 1941a: 138, Tab 6, Fig. 28, 35; 1941b: 201; 1042a: 383; 1944: 316, 321; 1945: 221; 1947: 266; Roewer 1954: 313. Porrima diver sa: Mello-Leitao 1947: 266. Porrimosa lagotis: Roewer 1959: 1002, 1005. Notas.— i) No se considera correcta la inclusion de Ocyale lagotis Holmberg, 1876 en la sinonimia de esta especie ( O . lagotis es un nomen nudum (ICZN, Art. 12); los tipos de Holmberg estan perdidos. ii) La especie es valida a partir de 1941 cuando Mello-Leitao describio, dibujo el epigino y dio la fotografia de una hembra. iii) Se designo lectotipo a la hembra N° 14945 (ICZN, Art. 74) depositada en la coleccion del Museo de La Plata. Diagnosis.— Esta especie difiere morfologicamente de P. castanea por los caracteres antes anotados en la diagnosis de aquella: epigino, mas chico, y espermatecas. La colora- cion y el diseno dorsal del abdomen, son significativamente distintos. Macho.— Largo total: 10.8 a 13.8 mm. Cefalotorax: largo 5.5 a 7.0 mm; ancho 3.7 a 4.6 mm (6 especimenes medidos); hordes laterales castano; mancha submarginal continua amarillo claro; 6 lineas finas radiales castano convergentes hacia el centro del surco 152 THE JOURNAL OF ARACHNOLOGY toracico. Esternon: naranja amarillo. Queliceros: castano naranja. Patas: amarillo, meta- tarsos y tarsos castano naranja, metatarso y tarso de pata I escopulados. Abdomen: dorsal castano verde, sin diseno; areas laterales diseno reticulado castano verde sobre fondo amarillo claro. Palpo como los esquemas de las figuras 6-8. Hembra.-Largo total: 10.0 a 18.5 mm (15.35 ± 2.18). Cefalotorax: largo 5.1 a 8.3 mm (7.02 ± 0.83), ancho 4.0 a 5.9 mm (5.20 ± 0.53) (21 especimenes medidos). Estruc- tura general semejante al macho. Cefalotorax: igual al macho pero de tinte mas oscuro. Esternon: castano naranja. Queliceros: rojo castano. Patas: castano oscuro. Abdomen: una franja mediana castano oscuro cuyo diseno es semejante al esquema de la figura 11; areas laterales salpicadas de puntos castano; ventral una franja mediana castano bordeada Figs. 6-10.— 6-8: Porrimosa lagotis (Mello-Leitao); 6, palpo derecho (Maldonado, Uruguay); 7, apofisis (Maldonado, Uruguay); 8, apofisis (N° 14944, MLP, Santa Fe, Argentina. Det. Mello-Leitao) vista lateral. 9-10: Porrimosa harknessi (Chamberlin); 9, palpo derecho (N° 269; MCZ) holotipo; 10, apofisis (N° 269; MCZ). a. 1. c. = apofisis lateral del conductor, a. m. = apofisis mediana, C. = cymbium, c. = conductor, e. = embolo, h. b. = haematodocha basal, t. = tegulum. CAPOCASALE— ESPECIES DEL GENERO PORRIMOSA 153 de puntos amarillo (Fig. 12). En el 27% de los especimenes observados no existe dicha franja, siendo unicamente amarillo. Epigino y espermatecas como los esquemas de las figuras 3-5. Comentarios.— En la description anterior se incluyen especimenes machos y hembras nacidos en el laboratorio. Criados desde huevo hasta adultos vivieron 8-13 mesesy cum- plieron 9-11 mudas (12 especimenes estudiados). La coloracion de los especimenes inmaduros, vivos, no es diferente de la de los adultos. La region mas distinta es el cefalototax, rojo naranja; el abdomen tiene identico diseno que en los adultos aunque es gris verdoso; las patas son gris amarillo. Dicha coloracion se mantiene hasta la penultima muda, luego de la cual llegan a adulto y cambian los tintes. Habitat— Los especimenes recolectados en Uruguay se hallaron sobre los 25 cm del suelo, cerca de piedras o sobre Eryngium sanguisorba Cham., en tubos de seda como lo describieron Simon (1898) y Vellard (1936). Un especimen de Argentina, Misiones, fue colectado en un tubo de seda hecho sobre un arbusto de 2 m de altura. Localidades validas — Argentina: Santa Fe, Guadalupe, Cordoba, Alta Gracia. (Demas indicaciones hechas por Mello-Leitao, se apoyan en especimenes inmaduros). Brasil: Parana; Curitiba. Uruguay: Maldonado: Sierra de la Ballena, Cerro Sension, Cerro Pan de Azucar; Tacuarembo: Tambores; Lavalleja: Arequita, Aguas Blancas, Marmaraja; Canelones: Marindia; Salto: El Espinillar. Distribution conocida— (Mapa 1) Centro de la Rep. Argentina, sur de la Rep. Fed. Brasil y Rep. 0. del Uruguay. Comportamiento sexual.- Re sumen (Costa, comunicacion personal). Cortejo: el macho avanza lentamente sobre la tela horizontal con retrocesos bruscos y patas anteriores elevadas. Los avances se alternan con periodos donde, sin adelanto, el macho realiza: 1) con las patas extendidas, flexiones intermitentes de tipo convulsivo, 2) tamborileo suave y esporadico con los palpos y 3) quietud. Lentamente el macho se aproxima al tubo, la hembra emerge y lo traba. Si es receptivo, retrocede trabando o rehuye para trabar Figs. 1 1-12. -Esquemas mostrando disenos abdominales: 11, Ponimosa lagotis (Mello-Leitao) dor- sal; 12, Porrimosa castanea (Mello-Lietao) ventral. El 78% de los especimenes hembras observados, de Porrimosa lagotis, tienen diseno ventral semejante al de la figura 12. 154 THE JOURNAL OF ARACHNOLOGY nuevamente mas adentro del tubo y permitir la monta. (8 parejas observadas). Copula : siempre dentro del tubo o en la entrada. Position tipica de los licosidos (Gerhardt II). Hay eyaculaciones multiples de cada palpo, alternandose en su uso. Hacia el final, solo una o dos eyaculaciones por palpo. En cada una hay erection de espinas y cese de las vibraciones del abdomen del macho (3 copulas observadas). Dos hembras reaccionaron provocando la huida del macho (duration de cada copula: 38.9 y 43.4 min). Otra copula en la que el macho se retiro sin violencia duro 87.8 min. Material examinado (especimenes determinados por Mello-Leitao). -Diez especimenes, de: Cordoba (1 hembra; 4 inmaduros); Tucuman (1 hembra inmadura); Salta (3 inmaduros); Santa Fe (1 hembra), Argentina, depositados en el Museo de La Plata. Dos especimenes de Corrientes, Argentina (1 hembra inmadura; 1 inmaduro), depositados en el Museo de La Plata. Dos especimenes de Curitiba, Brasil (2 hembras), depositados en el Museu Nacional de Rio de Janeiro. Localidades nuevas (especimenes determinados por el autoi). -Argentina: Santa Fe. Silva, sin fecha (Biraben), 1 macho (Museo de La Plata); Misiones. Cataratas Iguazu. 12 mar. 1976 (F. Costa), 1 hembra (Museo de Historia Natural, Montevideo). Uruguay: Lavalleja. Aguas Blancas, 30 set. 1979 (F. Perez, J. Basso), 2 hembras (MHNM), Marmaraja, 12 may. 1962 (C. Carbonell, M. Monne), 1 hembra (MHNM), Sierra de la Ballena, 29 ago. 1976 (F. Costa, M. Urruti), 1 macho, 3 hembras (MHNM), Cerro Arequita, 6 ago. 1978 (F. Costa, F. Perez) 2 hembras (MHNM); Salto. El Espinillar, 16 ago! 1979 (R. Capocasale, F. Costa), 10 hembras (MNHM); Maldonado. Cerro Pan de Azucar, 3 set. 1978 (F. Perez, J. Basso), 5 machos, 11 hembras (MHNM), Sierra de la Ballena, 29 ago. 1976 (F. Costa, M. Urruti), 1 macho, 3 hembras (MHNM), Cerro Sension, 28 may. 1978 (F. Perez, J. Basso) 1 hembra (MHNM); Canelones. Marindia, 16 jul. 1978 (F. Costa, M. Urruti), 1 hembra (MHNM); Tacuarembo. Pozo Hondo, 4 set. 1971 (F. Achaval), 1 hembra (MHNM). Porrimosa securifera (Tullgren, 1905), nueva combination Lycosa securifer Tullgren 1905: 66, Tab. 8, Fig. 32; Bonnet 1957: 2663 (securifera). Isohogna securifer: Roewer 1954: 262. Comentarios.— Holotipo de Moreno, Jujuy, Argentina (1 hembra inmadura) exami- nado, depositado en el Naturhistoriska Riksmuseet. Se considera species inquirenda. La ubicacion de la especie bajo el genero Isohogna debe estimarse errada. La mention hecha por Chamberlin (1916) debe ser eliminada de la sinonimia de acuerdo a lo siguiente: Lycosa securifer: Chamberlin, 1916: 282. Dos Figs. 13-14.-13: Queliceros de Porrimosa lagotis (Mello-Leitao) (Canelones, Uruguay) vista ventral (El 97% de los casos estudiados tuvieron 3 dientes en el borde posterior interno (flecha) de sus queliceros); 14, Oculario de Porrimosa castanea (Mello-Leitao). CAPOC ASALE- ESPECIES DEL GENERO PORRIMOSA 155 especfmenes de: Urubamba (1 hembra inmadura) examinada; Cuzco (1 hembra) identifi- cation erronea, Peru, depositados en el Museum of Comparative Zoology, Harvard Uni- versity. AGRADECIMIENTOS Mi sincere agradecimiento a las personas mencionadas a continuation: Dr. O. Francke, Prof. M. E. Galiano, Dr. W. J. Gertsch, Dr. B. J. Kaston, Lie. A. Mones y Dr. W. B. Peck, por ayudarme con sus comentarios y sugerencias. Ademas contribuyeron enviandome especfmenes en prestamo o autorizandome a revisarlos: Lie. R. Arrospide, Museo de La Plata, Argentina; Dr. H. Hubert, Museum National d’Histoire Naturelle, Francia; Dr. T. Kronestedt, Naturhistoriska Riksmuseet, Suecia; Dr. H. Levi, Museum of Comparative Zoology, Harvard University, U.S.A.; Prof. S. Lucas, Instituto Butantan, Brasil; Dr. E. Maury, Museo Argentino de Ciencias Maturates B. Rivadavia, Argentina; Dr. E. Taylor, Hope Department of Entomology, University Museum, Inglaterra; Prof. Ana TimoteoDa Costa, Museu Nacional de Rio de Janeiro, Brasil; Dr. F. R. Wanless, British Museum (Natural History) Inglaterra. Asimismo agradezco al Prof. F. Costa por permitirme publicar su resumen sobre el comportamiento sexual de Porrimosa lagotis. BIBLIOGRAFIA CITADA Bonnet, P. 1958. Bibliographia Araneorum 2 (4): 3027-4230. Douladoure. Toulouse. Brady, A. R. 1962. The Spider Genus Sosippus in North America, Mexico, and Central America (Araneae, Lycosidae). Psyche, 69 (3): 129-164. Bucherl, W. y S. Lucas. 1972. Sobre a posigao sistematica de Porrima callipoda Mello-Leitao, 1934 (Araneae, Lycosidae). Mem. Inst. Butantan, 36:267. Chamberlin, R. V. 1916. Results of the Yale Peruvian Expedition of 1911: The Arachnida. Bull. Mus. Comp. Zool. Harvard, 60 (6): 177-299. Gerhardt, U. 1924. Weitere Studien iiber die Biologie der Spinnen Arch. Naturgesch., 90:85-192. Holmberg, E. L. 1876. Aracnidos argentinos. An. Agr. Argentina, 4:1-30. International Commission on Zoological Nomenclature (I.C.Z.N.) 1964. International Code of Zoolog- ical Nomenclature, Suffolk. England. Richard Clay and Co. Ltd. 176 pag. I.C.Z.N. 1974. Amendments to the International Code of Zoological Nomenclature adopted since the IVth. International Congress of Zoology. Washington, 1963. Bull. Zool. Nomencl. 31 (2):77-89. Keyserling, E. 1877. Uber amerikanische Spinnearten der Unterordnung Citigradae. Verh. zool.-bot. Ges. Wien, 26:609-708. Keyserling, E. 1891. Die Spinnen Amerikas (Brasil. Spinn.), 3:1-278. Niirnberg. Mello-Leit2o, C. de. 1926. Algumas Aranhas do Brasil Meridional. Bol. Mus. Nac. Rio Janeiro, 2 (6): 1-18. Mello-Leitao, C. de 1934. Tres aranhas novas nas collegoes de Instituto Butantan. Mem. Inst. Butan- tan, 8:401-407. Mello-Leitao, C. de. 1941a. Los aranas de|cordoba', La Rioja, Catamarca, Tucuman, Salta y Jujuy. Rev. Mus. La Plata, (N.S.) (zool.), 2 (12):99-198. Mello-Leitao, C. de. 1941b. Aranas de la provincia de Santa Fe colectadas por el Prof. Biiaben. Rev. Mus. La Plata, (N.S.) (zool.), 2 (1 3): 199-225. Mello-Leitao, C. de. 1942a. Aranas del Chaco y Santiago del Estero. Rev. Mus. La Plata, (N.S.) (zool.), 2 (16): 381-426. Mello-Leitao, C. de. 1942b. Cinco aranhas novas do Peru. Rev. Brasil, Biol., 2 (4):429-434. Mello-Leitao, C. de. 1943. Catalogo das aranhas do Rio Grande do Sul. Arq. Mus. Nac. Rio Janeiro, 37:147-245. Mello-Leitao, C. de. 1944. Aranas de la Provincia de Buenos Aires. Rev. Mus. La Plata, (N.S.) (zool.), 3 (24): 31 1-393. Mello-Leitao, C. de. 1945. Aranas de Misiones, Corrientes y Entre Rios. Rev. Mus. La Plata, (N.S.) (zool.), 4 (29):213-302. 156 THE JOURNAL OF ARACHNOLOGY Mello-Leitao, C. de. 1947. Aranhas do Parana e Santa Catarina, das cole^oes do Museu Paranaense. Arq. Mus. Paranaense, 6 (6): 23 1-304. Mello-Leitao, C. de 1948. Contribuigao ao conhecimento da fauna araneologica da Guianas. An. Acad. Brasil. Sci., 20 (2): 15 1-196. Mello-LeitSo, C. de. 1949. Aranhas da Foz do Koluene. Bol. Mus. Nac. Rio Janeiro, (N.S.) (zool.), 92:1-19. Neave, S. A. 1940. Nomenclator Zoologicus. Vol. Ill, 1065 pag. Zoological Society of London. Pickard-Cambridge, O. 1877. On some new species of Araneida, with characters of two new genera and some remarks on the Families Podophthalmides and Dinopides. Proc. Zool. Soc. London, pp. 557-578. Roewer, C. F. 1954a. Katalog der Araneae von 1758-1954. Verlag von Natura. Bremen, 2a: 1-923. Roewer, C. F. 1954b. Exploration du Parc National de l’Upemba. Araneae Lycosaeformia, I. 1-420. Bruxelles. Roewer, C. F. 1959. Exploration du Parc National de l’Upemba. Araneae Lycosaeformia, II (Lycosi- dae). 519-1040. Bruxelles. Roewer, C. F. 1961. Uber Namen der Gattungen und Arten der Lycosidae (Araneae). Bull. Inst. Sci.nat. Belgique, 37 (8): 1-19. Simon, E. 1898. Histoire naturelle des Araignees. 2(2): 193-380. Paris. Taczanowski, L. 1873. Les Araneides de la Guyana fran5aise. Horae Soc. ent. Rossicae, 10:56-115. Tullgren, A. 1905. Araneida from the Swedish Expedition through the Gran Chaco and the Cordille- ras. Ark. Zool., 2 (19):1-81. Vellard, J. 1936. Le venin des Araignees. 1-312. Masson. Paris. Monographies de l’lnstitut Pasteur. Manuscript received April 1981, revised September 1981. Francke, O. F. 1982. Birth behavior in Diplocentrus bigbendensis Stahnke (Scorpiones, Diplocentri- dae). J. ArachnoL, 10:157-164. BIRTH BEHAVIOR IN DIPLOCENTRUS BIGBENDENSIS STAHNKE (SCORPIONES, DIPLOCENTRIDAE) Oscar F. Francke Departments of Biological Sciences and Entomology Texas Tech University, Lubbock, Texas 79409 ABSTRACT Four female D. bigbendensis gave birth in the laboratory. One parturition was normal and some of the young were reared to maturity, whereas the other three parturitions were abnormal and resulted in significant cannibalism of the newborn by their mothers. In the normal parturition 37 young were born during about 30 hours. The young emerged tail-first and upside-down, and were free of any “birth membranes.” It is hypothesized that the tail-first emergence of the young is directly related to the katoikogenic type of embryonic development (in embryonic diverticula) that these scorpions undergo. The average birth lasted 6.9 ± 4.0 minutes, and the average interval between consecutive births was 44.4 ± 30.8 minutes. The relatively large standard deviation is due to the emergence of the young mostly in pairs separated by considerably longer intervals between pairs. It is hypothesized that the paired births are related to the bilateral symmetry of the female reproductive system. INTRODUCTION The reproductive biology of diplocentrid scorpions is poorly known. Certain aspects of the behavior of Nebo hierichonticus (Simon)(Nebinae), such as mating and parturition, were studied in Israel by Shulov et al. (1960) and by Rosin and Shulov (1963). The New World is inhabited by six genera of diplocentrids, all belonging to the subfamily Diplo- centrinae. Williams (1969) observed parturition by Diplocentrus whitei (Gervais), but gave no specifics in his report. Thus, the only published observations on the reproductive biology of diplocentrine scorpions are those I made on Diplocentrus spitzeri Stahnke from Arizona (Francke 1981). My objectives here are to describe and analyze the birth behavior of Diplocentrus bigbendensis Stahnke from Texas. MATERIALS AND METHODS A sample of 27 D. bigbendensis , including seven adult females, were collected on 17 June 1974 in Big Bend National Park, Brewster Co., Texas. The specimens were placed individually in 1 1.5 X 10.5 X 7 cm plastic containers with a moist piece of paper towel on the bottom, and transported to Tempe, Arizona. There a 2-3 cm layer of soil was added to the containers, which were then kept in darkness at room temperature (approx. 25° C). The scorpions were offered live crickets and water once a week. On 10 September 1974 one female was in the process of giving birth, and a second one had assumed a similar “parturition” posture. Following the initial disturbance with 158 THE JOURNAL OF ARACHNOLOGY “white” light a change was made to red light to minimize further disturbance. Four parturitions occurred in the laboratory, three in September and one in November 1974. One parturition (Female E) was observed without interruptions from start to finish and is considered to have been normal because (a) no apparent complications were detected, and (b) the young have been successfully reared to maturity (Francke, in preparation). Partial observations were made on the other three parturitions (Females A, B, and D), which are considered abnormal because in each case the mother attempted to cannibalize the young and prevented them from climbing upon her back. The normal parturition is described first, followed by remarks and comparisons with the three abnormal parturi- tions. FEMALE E Pre-partum behavior.— Sporadic observations from 10-17 September 1974 showed that this female was the most active, engaging in digging behavior frequently. Since this behavior has not been previously described for any diplocentrid scorpion, a summary follows. Digging behavior consisted of two phases: digging with the legs and tail-scraping. Digging with the legs usually started with the front two legs on one side, flexed under the body, loosening soil with the ventral spines of tarsomere II and shoving it posterolater- ally. As soil accumulated along the side of the mesosoma the third and fourth legs on that side were brought into action, kicking the soil backwards by forceful extension of the joints. Often the actions of the two ipsilateral pairs of legs were recurrently alternated so that a stroke of soil-loosening and shoving by legs I and II was immediately followed by a stroke of soil-kicking by legs III and IV. After one bout of 20-25 cycles a moderate pile of soil accumulated to one side of the female at the level of the basal segment of the metasoma. The female would cease digging, move forward slightly, and engage in tail- scraping. This phase started with the metasoma moderately curled and lying sideways on the side of the soil pile. The metasoma would then be lifted off the substrate, moved sideways beyond the pile of soil, returned to the substrate, and slightly extended while being returned to an axial position in a swiping motion. With each swipe the top layer of the soil pile was removed and shoved backwards and medially. Eight to ten swipes were required to level the soil, at which time the female resumed digging with the legs on the contralateral side. Parturition— On 18 September 1974 at 0931 hrs. female E assumed the parturition posture with her metasoma backed against a corner of the container. The prosoma was lifted off the substrate with the carapace forming an angle of approximately 30° to the substrate. The mesosoma was moderately arched, with sternites IV and V resting on the substrate. The metasoma was strongly flexed dorsomedially. The pedipalps were moder- ately flexed and the chelae were parallel to the body axis with the fingers wide open. The first pair of legs were strongly flexed under the prosoma, crossing each other medially and forming the birth basket. The second pair of legs were slightly flexed with the tarsi resting on the substrate submedially, and during the initial births did not form a functional part of the birth basket. The female proceeded to give birth to 37 young during the following 27 hours. Table 1 is a chronological summary of the events and behaviors recorded during this parturition. The posture of the female during the first few births was as described above, and only one change was noticed during the births of the young. Between births the female’s FRANCKE-BIRTH IN DIPLOCENTR US BIGBENDENSIS 159 Table 1. -Chronology of parturition by Diplocentrus bigbendensis Stahnke, female E and behavior of the young. Times are reported using the 24-hour system, r = female’s right side, 1 = female’s left side. BIRTH ASCENT TO FEMALE ORDER DATE (Sept.) TIME APPEAR TIME DROP RANK DATE (Sept.) TIME UP SIDE UP 1 18 0937 0939 7 18 1615 r 2 18 0947 0951 2 18 1323 r 3 18 1012 1019 1 18 1230 r 4 18 1024 1026 31 19 1046 1 5 18 1029 1045 4 18 1426 r 6 18 1114 1127 3 18 1410 r 7 18 1209 1212 11 18 2007 1 8 18 1237 1245 5 18 1510 r 9 18 1259 1311 9 18 1800 1 10 18 1408 1412 17 19 0004 1 11 18 1452 1500 6 18 1605 r 12 18 1507 1516 8 18 1741 r 13 18 1547 1554 13 18 2015 1 14 18 1650 1655 18 19 0004 r 15 18 1709 1720 20 19 0035 r 16 18 1800 1809 10 18 1922 1 17 18 1819 1825 12 18 2010 r 18 18 1915 1918 22 19 0147 r 19 18 1925 1943 14 18 2156 1 20 18 2041 2043 15 18 2218 1 21 18 2046 2049 16 18 2305 r 22 18 2115 2122 23 19 0227 1 23 18 2224 2228 19 19 0035 r 24 18 2230 2233 24 19 0330 1 25 18 2321 2329 21 19 0046 1 26 19 0004 0008 37 19 1639 1 27 19 0026 0030 25 19 0437 r 28 19 0231 0236 36 19 1446 r 29 19 0317 0322 26 19 0520 r 30 19 0335 0343 32 19 1217 1 31 19 0426 0435 27 19 0600 1 32 19 0520 0522 29 19 0948 1 33 19 0642 065 3 28 19 0835 r 34 19 0750 0758 34 19 1412 1 35 19 0844 0850 30 19 1016 r 36 19 1002 1013 33 19 1217 r 37 19 1215 1222 35 19 1446 r pectines were held at an angle of about 15° to the body axis (or about 45 from the substrate). During births the pectines were extended, assuming a position perpendicular to the substrate. Thus, it was possible to anticipate the emergence of each young by the female’s behavior. At 1055 hrs., after five young had been born, the female’s second pair of legs was recruited to form an intergral part of the birth basket. Shortly before the seventh birth, the female engaged in a short bout of pedipalp movement. The right and left pedipalps were alternately shifted back-and-forth. The pedipalp chelae were then moved from a 160 THE JOURNAL OF ARACHNOLOGY subparallel to a transverse position with respect to the body axis, and the fixed fingers almost touched along the midline in front of the carapace. A few minutes before the eighth and ninth births, respectively, the female briefly raised the pedipalps off the substrate while simultaneously “pushing” with the metasoma against the corner of the container. After the ninth birth the female assumed a more relaxed position: the body was not as strongly angled with the substrate, the pedipalp chelae were held convergently at about 45° to the body axis and the fingers were closed. The parturition position, with pedipalp chelae transversely across the midline and the right and left fixed fingers almost touching, was resumed three minutes before the tenth birth. The tenth through twelfth births were preceded by pedipalpal and metasomal movements as described above; and three minutes after the twelfth birth the female reverted to a relaxed posture. At 1536 hrs. stronger pedipalpal flexions occurred and the female resumed the parturi- tion position. The mother showed a noticeable change in behavior as the thirteenth young emerged: the prosoma was suddenly and strongly arched backwards, and the pedipalps were extended and stretched up-and-back quite forcefully (“throwback” behavior). The female relaxed after the young emerged, and shortly before each of the following six births she resumed the parturition position after some pedipalpal and metasomal motion. The movements by the female increased in frequency and intensity from the twentieth birth on. Prior to resuming the parturition posture the female would strongly arch back- wards, raising the prosoma off the substrate. The pedipalps would be partially extended and raised off the substrate, and the female would then alternately twist the anterior half of the body first to one side of the midline and then to the other while stretching the pedipalps up-and-back in 3-5 consecutive cycles. The vigorousness of the female’s motions, their timing with respect to the births, and their increase in intensity and frequency with progressively later births suggest that they may serve to stimulate and facilitate the passage of the young out of their respective ovariuteral diverticula and forward along the ovariuterus to the genital opening. The female also assumed a relaxed posture between births from about the twentieth birth on. The relaxed posture was characterized by having the pedipalp chelae fingers !4 open, the chelae divergent from each other, the prosoma lowered, and the telson positioned over tergites III— IV. The parturition posture was usually assumed a few minutes before each birth: the pedipalp chelae fingers were wide open, the chelae were convergent to each other, the prosoma was raised (by strongly arching the mesosoma), and the telson was held over tergites V-VI. The female abandoned the parturition position and “broke” the birth basket at 1448 hrs. on 19 September, approximately 29 hours after she had assumed it and 2.5 hours after the 37th birth. The average interval between consecutive births (time of initial appearance) was 44.4 ± 30.8 min. (mean ± S. D., n = 36). The shortest interval was 5 min.; and the longest interval was 133 min. The elapsed time between the appearance of the first and the last young was 26 hrs. 38 min.; and the time from appearance of the first young until the last one had settled on the female’s back (#26) was 31 hrs. 02 min. On 19 September the female engaged in periodical bouts of throwback behavior until about 1930 hrs., or for about seven hours after the last birth. Post-partum behavior— The female fed on a live cricket on 20 September, and ate another one on 24 September. She also cannibalized four young before they could molt and disperse. The female died of unknown causes in March 1975. Behavior of the young— All the young were born tail first and free of any enclosing membranes. Two of the young appeared right-side up, and 35 appeared upside-down. The FRANCKE- BIRTH IN DIPLOCENTR US BIGBENDENSIS 161 emergence of each young, from the time the telson could be observed coming out to the time the young dropped into the birth basket, lasted 6.9 ± 4.0 min. (n = 37). The fastest were four births that required 2 min. each, and the longest one took 18 min. As they emerged the young would flex and stretch their appendages, and soon after being com- pletely out, they could move on their own. The side of the female along which a newborn climbs is apparently randomly deter- mined: 21 young went up the female’s right side and 16 went up her left side (Table 1). The time elapsed between completing emergence and setting on the mother’s back was 295.6 ± 282.9 min. (Table 1). The fastest young (#11) took only 65 min., while the slowest (#4) required 24 hrs. 20 min. On the night of 19 September 1974 the young, having just climbed upon their mother, were randomly positioned on her dorsum. Gradually the young shifted around with respect to each other, and by 3 October a definite pattern appeared. The young, 2-3 layers deep, were positioned transversely across the mother’s dorsum, facing each other medially. A narrow gap was evident along the female’s midline. Three young were down on 4 October 1974 actively moving around in the vicinity of their mother. The young molted to the second instar on 6-7 October 1974, and the last one left the mother on 16 October. In order to molt the first-instar young assumed a characteristic posture unlike that seen in subsequent molts. The body is strongly arched dorsally, and the metasoma is nearly straight and pointing slightly ventrad due to the arching of the body. The pedipalps and legs are flexed and crossed under the body— perhaps aiding in arching it— and the chelicera are fully extended. OTHER FEMALES Three other females gave birth in the laboratory. None can be regarded entirely normal or successful, but each offers some glimpses of behavior that corroborate the patterns described for female E. Female A.— On 10 September 1974 at 1540 hrs. this female was found to have as- sumed the parturition posture, and the telson of the first young was emerging from her SEQUENTIAL BIRTH INTERVALS Fig. 1. -Interval (in minutes) between the initial time of appearance of consecutive births by Diplocentrus bigbendensis Stahnke female E in the laboratory. 162 THE JOURNAL OF ARACHNOLOGY genital opening. The posture assumed by female A was very similar to that described for female E, and the first young was emerging tail-first as those of litter E. The female was immediately transferred to a bigger container to facilitate observations and photography. The disturbance appeared to affect all subsequent events: the first young did not com- plete emergence until 25 hours later, and then eight additional young were born in less than 80 min. (1945 hrs. to 2105 hrs., 1 1 September). Three pairs emerged almost simul- taneously; in each case one member of a pair emerged tail-first and the other emerged head-first. The nine young that were born in rapid succession were coated with a viscous substance and stuck to each other, thus being unable to stretch their appendages or to climb onto their mother’s back. Eventually the female abandoned the parturition posture and attempted to cannibalize the young. This female produced an additional 29 young over the next five days, as follows: 23 on 12 September, 4 on the 13th, none on the 14th, and one each on the 15th and 16th, respectively. Many of the births occurred late at night and were not observed; masses of young stuck to each other and on two consecutive mornings were found hanging from the female’s genital opening. The births that were observed occurred mostly in pairs (i.e., two young born almost simultaneously followed by a time gap), and the newborn stuck to each other and to members of other pairs. Between the births of each pair of young the female engaged in bouts of throwback behavior. To summarize the significance of these observations: female A had 38 young in six days (although 36 of them were born within a 36 hour period); the parturition posture and throwback behavior were similar to those observed in female E; and the majority of the births occurred in pairs. All the young died as first-instars, and the female died on 10 January 1975. Female B.-On 10 September 1974 at 1540 hrs. this female was found in the parturi- tion posture, carrying eight young on her back, and with 17 more on the ground under and near her. The female was cannibalizing one young, and the remnants of three others were found in the immediate vicinity (making a total of 29). Fearing that the sudden exposure to light had caused the female to move, thereby disrupting the birth basket for the young, I carefully placed all the young on the female’s back and returned them to darkness. During the next two days the female had at least four more young; and she cannibal- ized (or attempted to) several young, often knocking down those that would climb onto her back. In the early afternoon of 11 September I noticed 10 young arranged trans- versely across the female’s back, facing each other medially. To prevent further canni- balism the first-instars were isolated and maintained at near saturation humidity levels. Six survivors began their first molt on 27 September: five died during or shortly after the molt, whereas the sixth survived to reach sexual maturity. The significance of these observations is fourfold: (a) female B had 33 young (or more?), which compares favorably with the 37 and 38 young delivered by females E and A, respectively; (b) the pattern assumed by the young on their mother’s back was very similar to that described for female and litter E; (c) the first molt occurred at an age of 17-20 days in both litters; and (d) the lone survivor represents, to my knowledge, the first scorpion successfully reared in isolation from shortly after birth. Female D.— This female gave birth on 2-3 November 1974, about 1.5 months after the other three females. This female cannibalized the young a few hours after each one of them was born, and she moved around in her container between births, scattering the young as she went. Detailed observations were made on very few births, and in those the young emerged tail-first as did the young of female E. FRANCKE-BIRTH IN DIPLOCENTR US BIGBENDENSIS 163 DISCUSSION The parturition posture of D. bigbendensis is very similar to that of D. spitzeri , differing only in the fact that in the latter the second pair of legs are an integral part of the birth basket from the beginning (Francke 1981), whereas in the former they are recruited into the birth basket after a few births have occurred. In N. hierichonticus the stance assumed by the gravid female is also similar to that of Diplocentrus spp., apparent- ly differing only in that the pedipalp chelae are held parallel to the body axis during parturition, rather than transverse or oblique to it (Rosin and Shulov 1963 :pl. 3; how- ever, those photographs may picture a female during a “relaxed” phase, in which case there would be no significant differences between the two genera). Throwback and twist behavior before each birth were not observed in D. spitzeri , but were very distinctive in D. bigbendensis. In N. hierichonticus conspicuous movements of the tail are reported (Rosin and Shulov 1963). In some cases tail movements were accom- panied by elevation of the pedipalps, but it is difficult to determine if these pedipalpal elevations are homologous to the throwback behavior of D. bigbendensis. The time from initial appearance to dropping into the birth basket was 6.9 ±3.9 min. (n = 37) in D. bigbendensis , 61.6 ± 54.2 min. (n = 8) in D. spitzeri , and “10 minutes on the average” in N. hierichonticus (Shulov et al. 1960). The young of D. spitzeri are relatively larger at birth (second-instar is 35% of female size) than those of N. hieri- chonticus (25%) and D. bigbendensis (22-23%) (Francke 1981). The passage of young through the genital opening of the mother may be determined by the young’s relative size: small young pass through considerably faster than larger young. The young of D. spitzeri and N. hierichonticus emerge tail first, as did those of D. bigbendensis from the normal birth (female E). Shulov et al. (1960) indicate that “In another case of parturition of N. hierichonticus the neonati were observed to emerge mostly in pairs, one with the head first, the other with the tail first.” Those authors did not indicate whether that was a normal and successful parturition or not; nonetheless, the same pattern occurred in D. bigbendensis female A births. A possible explanation for the tail-first emergence of the young in normal births in diplocentrids relates to their position during embryonic development. Diplocentrid scor- pions are katoikogenic, i.e., they develop in diverticula that arise perpendicularly from the main ducts of the ovariuterus. The embryos face the apices of the diverticula, and they literally have to back out (tail first) from the diverticula and make a 90° turn to enter the lumen of the oviduct. The path of least resistence would be to turn anteriorly with respect to the female because that is the direction towards the genital opening. Turning posteriorly would enable the young to enter the lumen, and then to move forward and be born head first. However, turning posteriorly where other young are awaiting their turn to come out would be mechanically more difficult than turning anteriorly. The relationship, if any, between abnormal births and some young emerging head first is unclear at this time. The overiuterus leads into paired oviducts which emerge into a common atrium at the level of the genital opening. Thus, it is possible for two young to move forward simul- taneously, one heading for each oviduct, and emerge as a “pair.” The pattern of two young being born in rapid succession occurs in normal as well as abnormal births (in the latter rapid succession becomes simultaneously, as reported for D. bigbendensis female A, and N. hierichonticus in one parturition). The intervals between consecutive births from litter E have been calculated from Table 1 and plotted in Fig. 1. The marked zig-zag 164 THE JOURNAL OF ARACHNOLOGY pattern, especially during the intermediate births, supports the explanation of two young moving forward at the same time. In D . spitzeri the pattern of intervals between consecutive births provides an indepen- dent test for the explanation derived from D. bigbendensis. The intervals in D. spitzeri were 128, 207, 45, 1065, 43, 202, 1513, and 180 min. respectively (Franc ke 1981). The median is 191, and the pattern of runs with respect to it is: below— above— below- above— below— above— above— below, deviating from a zig-zag only once. Intervals be- tween consecutive births are not known for N. hierichonticus (or any other katoiko- genous scorpion) and thus further testing of the hypothesis presented above cannot be offered here. In D. bigbendensis early births have shorter consecutive intervals than late births. The median in Fig. 1 is 44.5; in the first 12 births nine intervals are below the median, in the middle 12 births six intervals are below the median, and the last 12 births only three intervals are below the median. The pattern strongly suggests that the young nearest the genital opening, i.e., those that are in the anterior-most diverticula and have a shorter distance to travel in order to get out, are born first. The young farthest from the genital opening, which have to travel a greater distance to get out, are born last. ACKNOWLEDGMENTS This study was done while I was a graduate student in the Department of Zoology, Arizona State University, Tempe, and was supported by a Graduate School Fellowship. My special thanks to Dr. Mont A. Cazier for his guidance and encouragement, for his help collecting the specimens, for stopping me from preserving them prematurely (as my taxonomic instincts would have preferred), and for keeping food and coffee flowing in my direction during the long uninterrupted periods of observation. To the latter task Linda Draper, Jan Hubbard (now my wife), Charlie Moss, and Carlos Gomez also contri- buted significantly and I thank them. On some nights when I could no longer keep my eyes open, Charlie Moss took over the observations and his help in obtaining uninter- rupted records is deeply appreciated. Finally, James Cokendolpher, Gary Polis, and David Sissom helped in improving the manuscript with their constructive criticisms. LITERATURE CITED Francke, O. F. 1981. Birth behavior and life history of Diplocentrus spitzeri Stahnke (Scorpiones, Diplocentridae). Southwest. Nat., 25:517-523. Rosin, R. and A. Shulov. 1963. Studies on the scorpion Nebo hierichonticus. Proc. Zool. Soc. London, 140:547-575. Shulov, A., R. Rosin and P. Amitai. 1960. Parturition in scorpions. Bull. Res. Counc. Israel. 9B:65-69. Williams, S. C. 1969. Birth activities of some North American scorpions. Proc. California Acad. Sci., 37:1-24. Manuscript received February 1981, revised May 1981. Stevenson, B. G. and D. L. Dindal. 1982. Effect of leaf shape on forest litter spiders: Community organization and microhabitat selection of immature Enoplognatha ovata (Clerck) (Theridiidae). J. Arachnol., 10:165-178. EFFECT OF LEAF SHAPE ON FOREST LITTER SPIDERS: COMMUNITY ORGANIZATION AND MICROHABITAT SELECTION OF IMMATURE ENOPLOGNATHA OVATA (CLERCK) (THERIDIIDAE) Bruce G. Stevenson and Daniel L. Dindal Soil Ecology, State University of New York College of Environmental Science and Forestry Syracuse, New York 13210 ABSTRACT Effects of leaf shape and other habitat structure variables on spider community organization and on microhabitat selection of immature Enoplognatha ovata in deciduous forest litter were studied in central New York. Three Utter boxes of each of four litter treatments (curled maple leaves, flat maple leaves, curled filter paper disks, and flat filter paper disks) were sampled monthly from November 1979 to August 1980. Twelve similarly-sized samples of natural litter also were collected, and all Utter boxes and natural Utter samples were placed in Tullgren funnels for extraction of spiders. Spider species richness was significantly greater in curled Utter than in flat Utter. Significant differences in the composition of hunting spider families also were found. Differences in species richness and composition of hunting spiders was attributed to differences in habitat space, according to leaf shape. Each Utter treatment supported approximately the same proportion of hunting and web-building spiders and similar composition of web-building spider families. These data indicated similarity in prey resources among the Utter treatments. A three-way ANOVA for E. ovata density revealed significant treatment effects for leaf type, leaf shape, and month, and a significant interaction occurred between leaf type and month. More spiders were found in maple leaves than in filter paper disks and greater density occurred in curled Utter than in flat Utter. Preference of maple leaves and curled litter by this species may be due to increased amounts of habitat space in these microhabitats. Development of immature E. ovata, associated with increases in body size and greater needs for space, resulted in differential seasonal microhabitat selection. In spring and summer, spiders grew in size and Utter populations (both Utter boxes and natural Utter) decreased. At this time mature adults were found beneath tree leaves in the forest understory layer. INTRODUCTION For two decades, ecologists have been concerned with the relationship between habitat structure and faunal community organization (e.g., MacArthur and MacArthur 1961). In a study of deciduous forest Utter, Uetz (1974) used interstitial space between leaves as a measure of habitat space; curled or bent leaves produce greater interstitial space than do flat leaves. Litter depth and habitat space increase as the proportion of curled or bent leaves increases (Uetz 1974, Bell and Sipp 1975). 166 THE JOURNAL OF ARACHNOLOGY There is considerable evidence which suggests that habitat structure influences com- munity organization of spiders inhabiting deciduous forest litter. Within a guild of wan- dering spiders, species richness, diversity and equitability all were correlated with litter depth and habitat space (Uetz 1975). Increased microhabitat diversity tended to increase species richness of litter spiders since a greater variety of structural microhabitats were available to them. Jocque (1973) also found that spider species diversity increased with increased depth and complexity of litter. In addition, population density of litter spiders has been correlated with litter abundance and depth (Lowrie 1942, 1948, Berry 1967, Hagstrum 1970). Studies in which one or more of the variables of litter habitat structure are manipu- lated should give better understanding of the factors regulating spider species diversity within forest litter (Uetz 1975). For example, gradients of litter depth and associated habitat complexity have been studied under natural (Uetz 1976) and experimentally- modified (Uetz 1979) conditions. In both cases, wandering spider species richness in- creased with greater depth and complexity of litter. The relationship between species diversity and habitat structure depends, in part, on microhabitat selecton of individual species. Differential selection of microhabitats with different physical structures by potentially competing species may permit coexistence of those species (Enders 1974, Greenquist and Rovner 1976). Within forest litter, abundance of Lycosidae is greatest in shallow litter made up of flat leaves, and decreases with increasingly deep litter, containing more bent and curled leaves. Abundance of Club- ionidae, Gnaphosidae and Thomisidae increases as litter becomes deeper and more com- plex (Uetz 1976, 1979). Spiders in the latter two families use curled leaves for retreats (Kaston 1948, 1978). There are few analyses of the use of curled leaves by web-building spiders in forest litter or other habitats, although many species build silken retreats or egg sacs in curled leaves (Kaston 1978, Lubin 1978). Spiders inhabiting Mahonia aquifolium Nutt, largely occupied curled leaflets (Waldorf 1976). Curled and closely parallel leaflets provided more shelter than flat leaflets. This paper reports on the effect of altered leaf shape on community organization of forest litter spiders and on microhabitat selection of Enoplognatha ovata (Clerck) ( =Ther - idion redimitum [L.]) (Theridiidae). The objective of this study were: (1) to test the hypotheses: (a) species richness of litter-inhabiting spiders is greater in curled litter than it is in flat litter and (b) curled litter contains more spider guilds than flat litter, and (2) to determine: (a) if leaf shape influences microhabitat selection by immature E. ovata and (b) if its life history affects the choice of microhabitats. METHODS The study was conducted at the Soil Invertebrate Ecology Laboratory at the Lafayette Experimental Station of the SUNY College of Environmental Science and Forestry in Syracuse, New York. A field experiment was conducted in a mixed hardwood stand. Dominant vegetation consisted of red oak ( Quercus ruba L.) and sugar maple {Acer saccharum L.). Understory vegetation was composed of red oak, sugar maple, and black cherry ( Primus serotina Ehrh.). Two soil types present on this site were Cazenovia sandy loam and Ontario sandy loam (alfisols, glossoboric hapludalfs [USDA 1972]). Mull humus characterized the organic subhorizons; forest litter disappeared in about one year primarily due to earthworm activity (Pritchard 1941, Stevenson pers. obs.). STEVENSON AND DINDAL-LEAF SHAPE AND FOREST LITTER SPIDERS 167 Table l.-List of identified species extracted from litter boxes and the distribution of individuals in litter treatments. Those species listed as immatures were identified to morphospecies. SPECIES Flat Filter Curled Filter Flat Maple Curled Maple Paper Disks Paper Disks Leaves Leaves Family Theridiidae Enoplognatha ovata (Clerck) (= Theridion redimitum [L.] ) 14 37 56 114 Family Linyphiidae Bathyphantes pallida (Banks) 1 4 13 2 sp. 1 (immature) 3 7 1 3 sp. 2 (immature) 1 1 0 7 Family Agelenidae Circurina brevis (Emerton) 2 6 3 6 C. robusta Simon 2 2 3 7 C. sp 1 (immature) 1 0 3 1 C. sp. 2 (immature) 0 0 0 1 Family Hahnidae Neoantisetea agilis (Kerserling) 1 2 0 2 N. sp. 1 (immature) 0 1 0 1 Family Lycosidae Pirata sp. 1 (immature) 0 0 0 1 Family Anyphaenidae Anyphaena sp. 1 (immature) 0 4 0 5 Family Salticidae sp. 1 (immature) 1 0 1 1 TOTAL 26 64 79 152 Litter boxes were used to measure the influence of leaf shape on litter spider com- munity structure since they permitted modification of leaf shape prior to placement in the field (Stevenson and Dindal 1981). Litter boxes (10 cmx 10 cmx 2 cm) were constructed of metal hardware cloth (6.35 mm openings). Large wandering spiders, such as lycosids, may be excluded from the litter boxes because of the small openings. This possibility is unlikely, however, since no adult lycosids have been collected in pitfall traps at this study site (Stevenson, unpublished data; Tardiff and Dindal, unpublished). The largest wandering spider collected in pitfalls was Circurina robusta Simon which was also collected in litter boxes. Freshly-fallen sugar maple leaves were collected in October 1978. Leaves were wetted to saturation in distilled water and pressed flat in a plant press. Dried and flattened leaves were selected and separated into 60 batches of 10 leaves each. Similarly, 600 filter paper disks (9.0 cm in diameter, Whatman Company) were also separated into 60 batches of 10 disks each. Filter paper disks were chosen as a simulated leaf type to provide a non- nutritive substrate, of which the major variable would be leaf shape. Filter paper degrades primarily by weathering, not by biological decomposition processes, due to a high car- bon: nitrogen ratio (Dindal and Levitan 1977). Each filter paper disk and dried maple leaf was measured (± 0.01 cm2) on a LiCor Model 3100 Area Meter (Lamba Instrument Corporation, Lincoln, Nebraska). Surface area per leaf or disk was determined as the mean of four measurements. Average leaf or 168 THE JOURNAL OF ARACHNOLOGY Table 2. -Coefficients of correlation between spider species richness (number of species) and two habitat structure variables. Litter Treatment Variable November-March (13 df) April- August (13 df) Curled Maple Leaves Mean Leaf Size (B.E.)1 -.082 .001 Mean Leaf Size (A.E.)2 -.052 -.009 Mean Tube Opening Size .068 -.045 Number of Spider Webs .740** .417 Flat Maple Leaves Mean Leaf Size (B.E.) .025 .134 Mean Leaf Size (A.E.) -.131 .402 Mean Tube Opening Size n.a.3 n.a. Number of Spider Webs n.a. n.a. Curled Filter Paper Disks Mean Disks Size (B.E.) -.325 .174 Mean Disk Size (A.E.) .009 .309 Mean Tube Opening Size -.394 .687** Number of Spider Webs .191 .307 Flat Filter Paper Disks Mean Disk Size (B.E.) .232 -.073 Mean Disk Size (A.E.) .454 .288 Mean Tube Opening Size n.a. n.a. Number of Spider Webs n.a. n.a. 1 B.E. = before spider extraction. 2 A.E. = after spider extraction. 3n.a. = correlation analysis not appropriate. **p < .01. disk size per batch was calculated from the mean values recorded for 10 individual leaves or disks in each batch. Curled leaves and disks were made by wetting each disk or leaf in distilled water until pliable and then rolling it by hand to form a tube. Each batch was allowed to dry and then it was placed in a litter box. Although habitat space was not measured directly, curled litter provided more inter- stitial space than flat litter, on a qualitative basis, since flat litter formed tight layers and had a more compact appearance than curled litter. This compaction was especially evi- dent in flat filter paper disks since they lacked microrelief, such as venation, which provided interstitial spaces. A 10 m x 12 m grid with 1 m intervals was formed at the study site. Each litter box was randomly assigned to a position at the intersection of the grid intervals. All boxes were placed in the field by October 1, 1979; subsequent natural litter fall covered the boxes and was left in place. It partly helped to prevent rain damage to enclosed leaves or disks. Litter covering the boxes did not enter them and thus did not contribute to the litter treatments or influence litter structure. Twelve boxes, three of each litter treatment type, were randomly selected for removal in each month from November 1979 to August 1980 and were individually sealed in plastic bags to prevent loss of arthropods. A similarly-sized sample of natural litter was also removed from the area adjacent to each box and sealed in a plastic bag. A previous study (Stevenson and Dindal 1981) revealed that the sample size was adequate to detect differences in arthropod communities between litter treatments. Spiders in litter boxes STEVENSON AND DINDAL-LEAF SHAPE AND FOREST LITTER SPIDERS 169 Table 3. -Coefficients of correlation between Enoplognatha ovata density and three environmental variables. Litter Treatment Variable November-March April-August (13 df) (13 df) Curled Maple Leaves Mean Leaf Size (B.E.)1 -.288 .018 Mean Leaf Size (A.E.)2 -.155 .058 Mean Tube Opening Size -.006 .463 Flat Maple Leaves Mean Leaf Size (B.E.) .375 .099 Mean Leaf Size (A.E.) .225 .310 Mean Tube Opening Size n.a.3 n.a. Curled Filter Paper Disks Mean Disk Size (B.E.) -.234 .174 Mean Disk Size (A.E.) .011 .334 Mean Tube Opening Size -.394 .687** Flat Filter Paper Disks Mean Disk Size (B.E.) .245 -.050 Mean Disk Size (A.E.) .630* .083 Mean Tube Opening Size n.a. n.a. 1 B.E. = before spider extraction. 2 A.E. = after spider extraction. 3n.a. = correlation analysis not appropriate. *p < .05. **p < .01. and natural litter samples were extracted in large, steep-sided Tullgren funnels and fixed in a solution of isopropyl alcohol:glycerin:water (80:10:10). Contents of all boxes and litter samples also were hand-sorted for additional spiders. Cephalothorax widths of all E. ovata individuals extracted from litter boxes and natural litter samples were measured with a micrometer. These data were extrapolated to the nearest 0.1 mm. Following extraction, diameters of the two openings on each curled leaf or disk tube were measured to the nearest 1 .0 mm. The presence of spider webs within the tubs was scored (present or absent). Leaves or disks from each litter box were wetted and pressed flat in the plant press following hand-sorting for spiders. After drying, each disk or leaf was measured with the area meter as described previously. Three-way analyses-of-variance (ANOVA) were performed with leaf type (maple leaves or filter paper disks), leaf shape (curled or flat), and months (November through August) as independent variables. Dependent variables were spider species richness, equitability (J ; Pielou 1966), and density of E. ovata. The relationships between two habitat struc- ture indices (leaf or disk size and mean opening size of curled leaf or disk tubes) and species richness or E. ovata density was tested by product-moment correlation. Finally, differences in relative composition of spiders (guild composition and taxonomic families) were examined by tests of independence for Rx C contingency tables (G-test; Sokal and Rohlf 1969). RESULTS AND DISCUSSION COMMUNITY ORGANIZATION A total of 321 spiders were collected from litter boxes, representing 7 families and 13 species or morphospecies. A list of these species and the distribution of individuals is 170 THE JOURNAL OF ARACHNOLOGY given in Table 1. Those species listed as immatures were identified to the morphospecies level (see Stratton et al. 1979) since species-specific taxonomic characteristics (e.g., genitalia) either were not present on those organisms or could not be determined. Species Richness.— Spider species richness (number of species) in litter boxes was analyzed to evaluate the importance of leaf shape to community organization of litter spiders. Litter boxes containing curled litter had more spider species than did boxes with flat litter (ANOVA Fi,80 = 7.198, p < 0.01). Thus, the field experiment confirms the hypothesis that more spider species occur in curled litter. The amount of habitat space in litter of different shape influenced patterns of spider species richness. Since flat leaves and filter paper disks formed tight layers with reduced interstitial space, fewer species could co-exist in such microhabitats than in curled litter which provided more habitat space. These results agree with positive correlations between forest litter habitat space and wandering spider species richness (Uetz 1975, 1979). Correlations between species richness and habitat structure indices were calculated separately for each leaf treatment. They were also conducted for the first and second halves of the 10-month study period, since there are important seasonal differences in spider growth and metabolic activity (Moulder and Reichle 1972). All correlation coef- ficients were tested for statistically significant differences from zero by t-tests (Sokal and Rohlf 1969). Within boxes of maple leaves (both curled and flat), number of spider species was not significantly correlated with any of the variables measured (Table 2). Within boxes of LITTER TREATMENT Fig. 1. -Composition of hunting and web building spider in litter boxes of four litter treatments. Hunting and web building species represent the two major foraging strategies of spiders. (Litter treatments: FFP = flat filter paper disks, CFP = curled filter paper disks, FML = flat maple leaves, CML = curled maple leaves). STEVENSON AND DINDAL-LEAF SHAPE AND FOREST LITTER SPIDERS 171 curled filter paper disks, spider species richness was correlated (p < 0.01) with mean opening size of the filter paper tubes during the second half of the study period. Field observations indicated, at this time, that the ends of some filter paper tubes had been closed by rainfall impact and many individual tubes were completely flattened. This correlation suggests that spider diversity was limited by habitat space. Rain damage to these tubes caused reduction in habitat space which few species could exploit (see also Uetz 1975, 1976). Tubes of curled maple leaves retained their shape and thus access to tube interiors was unhindered. Finally, spider species richness in litter boxes containing flat filter paper disks was correlated (p < 0.05) with mean size of filter paper disks following arthropod extraction. This occurred during the first half of the experiment (Table 2). This correlation is spuri- ous since the size of filter paper disks exhibited extremely low variability. The coefficient of variation (CV - standard deviation/mean x 100; Sokal and Rohlf 1969) of disk size for flat disks was 0.075 ± 0.004 (x ± S.E.; n = 30) before placement in the field and 0.688 ± 0.337 after extraction of spiders. Moreover, all other correlations between leaf or disk size and number of spider species were not statistically significant. Thus, leaf size does not appear to be an important factor influencing forest litter spider community organiza- tion. Similarly, Reice (1980) noted that arthropods in litter habitats react more to inter- stitial space than to substrate area. Equitability.— Leaf shape had no effect on the equitability of spiders within species (ANOVA F i ,8 o = 2.956, 0.05 < p < 0.10). This result contrasts with data from Uetz (1975) who found that equitability of wandering spiders was correlated with both depth LITTER TREATMENT Fig. 2. -Composition of web building spiders by family in each of the four litter treatments. Litter treatment key given in Figure 1. 172 THE JOURNAL OF ARACHNOLOGY Table 4. -Monthly differences in mean density of Enoplognatha ovata in litter boxes. Differences in means were analyzed by Duncan’s Multiple Range Test. Those means not connected by the same underline are significantly different at the .05 level. Jul. Aug. Jun. May Feb. Apr. Nov. Jan. Apr. Dec. 0.08 0.08 0.17 1.08 1.67 1.91 2.50 2.92 2.92 4.08 and habitat space within forest litter. Uetz suggested that complex litter environments, which contain curled leaves, regulate spider species diversity by reducing differences in the relative abundance of species, in addition to increasing the number of species. Our results indicate that regulation of spider equitability in complex litter is not solely due to differences in leaf shape. Instead, increased survivorship of immature spiders in deep and complex litter, due to greater abundance of refuges from predation and cannibalism (Uetz 1975), may be the proximate cause of greater evenness in such habitats. Guild Composition.— Composition of spiders in the four litter treatment types was compared according to method of prey capture: hunting spiders or web-building spiders. Hunting spiders included wandering species (after Uetz 1975) and jumping species. Within the web-building group, only Linyphiidae and Theridiidae were collected. Only species of the genus Circurina (hunting spiders) were found in the Agelenidae. Similar percent compositions of spider functional groups were found in each litter treatment (Figure 1). Web-building spiders comprised 87.5%, 84.1%, 73.1%, and 76.6% of total spiders in flat maple leaves, curled maple leaves, flat filter paper disks, and curled filter paper disks, respectively. Percent composition of the two functional groups was not significantly different between litter treatments (G = 4.652, 0.1 < p < 0.5). Distribution of web-building spider families was also similar in the four litter treat- ments (Figure 2). Theridiid spiders comprised 80.0%, 89.8%, 73.7%, and 75.5% of web- building spiders in litter boxes of flat maple leaves, curled maple leaves, flat filter paper disks, and curled filter paper disks, respectively. Litter treatment did not affect relative composition of web-building spider families (G = 3.934, 0.1 < p < 0.5). Stratton et al. (1979) found similar proportions of hunting and web-building spiders on three coniferous tree species and suggested that there were similar basic resources (prey and spatial parameters) on all three trees. Since all Utter boxes were located at the same study site, the four litter treatments supported similar prey types for spiders (Stevenson 1980). However, spatial properties of the four microhabitats differed con- siderably because of differences in leaf shape. Differences in these spatial properties apparently caused important differences in com- position of hunting spider families (Figure 3). For example, Anyphaenidae were found exclusively in boxes of curled litter (20% of total hunting spiders in boxes of both curled maple leaves and curled filter paper disks). Salticids were more prominent in boxes of flat litter (10.0% of hunting spiders in flat maple leaves, 14.3% in flat filter paper disks) than in boxes of curled litter (4.0% of hunting spiders in curled maple leaves, 0.0% in curled filter paper disks). Litter treatment had a significant influence on the relative composition of hunting spider famiUes (G = 15.612, p < 0.005). Although relatively little is known about the ecology of litter-inhabiting anyphaenids, they are similar to clubionids in morphology and habit (Kaston 1948, 1978). Clubionids STEVENSON AND DINDAL-LEAF SHAPE AND FOREST LITTER SPIDERS 173 make tubular retreats in rolled-up litter leaves (Kaston 1948, 1978). Anyphaenids make similar retreats in the laboratory (Uetz pers. comm.) and may also build them in curled litter. Salticid spiders occurred more frequently in flat litter than in curled litter (Figure 3). These jumping spiders orient to their prey visually (Kaston 1948) and need a solid platform from which to jump (Stratton et al. 1979). Flat maple leaves and filter paper disks provide greater unobstructed surface area as a jumping platform than does curled litter. Further, visual orientation to prey is enhanced in flat litter. MICROHABITAT SELECTION OF IMMATURE ENOPLOGNATHA OVATA A total of 289 individual E. ovata were collected from litter boxes and natural litter samples. All spiders were sexually immature. Further, this species was the dominant spider in all litter treatments (see Table 1). Influence of Leaf Shape on Microhabitat Selection.— Relative abundance of E. ovata within litter boxes reflected microhabitat selection by this species. More spiders were found in curled litter than in flat litter (ANOVA, ,80 = 18.571, p < 0.001). These data are consistent with published reports on the use of curled leaves by E. ovata. This spider is one of several web-building species which inhabit curled leaflets on Mahonia aquifolium (Waldorf 1976). Egg sacs of E. ovata are usually attached to the underside of a curled leaf (Kaston 1948). Thus, there is considerable evidence that curled leaves are preferred microhabitats for E. ovata. Two hypotheses are suggested for preference of curled leaf microhabitats by this LITTER TREATMENT Fig. 3. -Composition of hunting spiders by family in each of the four litter treatments. Litter treatment key given in Figure 1. 174 THE JOURNAL OF ARACHNOLOGY species: (1) curled litter provides refuges from predation and hence contribute to in- creased survivorship of individuals inhabiting them, and (2) curled litter provides suit- able web sites due to architectural properties of the microhabitat and large amounts of habitat space (see Uetz 1974, 1975). Previous studies suggest or demonstrate that litter complexity, including the presence of curled leaves, can reduce inter- and intra-specific predation among litter spiders (Edgar 1969, Hallander 1970, Uetz 1976, 1979). Further, the availability of web substrate, coupled with sufficient habitat space for web construc- tion and function, are among the most important environmental factors limiting distribu- tions of web-building spiders (Lowrie 1948, Turnbull 1960, Duffey 1966, Enders 1973, 1974, Robinson 1981). Habitat space within forest litter increases as the percent of curled leaves increases (Uetz 1974, 1975, Bell and Sipp 1975). Our data are insufficient to completely discriminate between these two hypotheses for explanation of micro- habitat selection by E. ovata. However, during the first half of the study period, E. ovata density was correlated (p < 0.01) with the number of spider webs in tubes of curled maple leaves (Table 3). This correlation suggests that most of the spiders which inhabited these microhabitats built and used webs. Spatial properties of the leaf tubes no doubt were important for web construction. Maple leaves supported greater density of E. ovata than did boxes with filter paper disks (ANOVA, F^go = 31.101, p < 0.001). Differences in amounts of habitat space, according to leaf type, explain these results. Since maple leaves have surface microrelief (e.g. venation), adjoining leaves in the flat maple leaf treatment retain interstitial space, whereas flat filter paper disks compact completely. Thus, more spiders are able to occupy the greater habitat space in flat maple leaves. Fig. 4. -Monthly variation in density of Enoplognatha ovata in natural litter samples (x; n = 12). STEVENSON AND DINDAL-LEAF SHAPE AND FOREST LITTER SPIDERS 175 Spider density was correlated (p < 0.01) with mean open tubular diameter of curled filter paper disks during the second half of the experiment (Table 3). At this time, some curled filter paper tubes and their openings were compacted by rainfall which caused reductions in habitat space. Density of E. ovata was limited by this reduced habitat space. Curled maple leaves retained their shape and habitat space throughout the experiment. Hence, E. ovata was not prevented from colonizing curled maple leaves. Forest litter habitat space is a function of leaf size as well as leaf shape and litter depth. Greater interstitial space occurs between large leaves than between small ones. Thus, microhabitat selection of E. ovata may be related to leaf size. However, correlations between E. ovata density and mean disk or leaf size were not significant (Table 3). Thus, leaf size does not appear to influence its microhabitat selection. Influence of Life History on Microhabitat Selection.— There were significant seasonal differences in E. ovata abundance (ANOVA, F9,80 = 7.190, p < 0.001). Months in which there were population density maxima in litter boxes (November through January and April) were different (p < 0.05) from all other periods (Table 4). A significant interaction occurred between leaf type and month (ANOVA, F9,80 = 2.802, p < 0.01), which indicates that E. ovata density was more variable in maple leaves than it was in filter paper disks. Further, E. ovata density in natural litter was highest in January and in April to May (Figure 4). These months correspond to the times density maxima occurred in litter boxes. Thus, these monthly differences in E. ovata density reflect differential micro- habitat selection by season. E E 0 1 X < ec O x H- o mJ < u Fig. 5. -Monthly variation in cephalothorax width of Enoplognatha ovata in litter boxes and natural litter samples (x ± SE; n = 12). 176 THE JOURNAL OF ARACHNOLOGY Two other pieces of evidence reveal that changes in ontogeny of this species influence its microhabitat selection. First, cephalothorax widths from spiders extracted from litter boxes were compared to cephalothorax widths from spiders from natural Utter (Figure 5). Body size of these spiders remained essentially constant in both natural litter and Utter boxes from November through April. From May to August, spider body size increased, although spiders were absent from Utter boxes in July and from natural Utter in August. Thus, growth of this spider occurred in the second half of the experiment, particularly from May to June. Second, during summer months (particularly June and July), many large, sexually mature E. ovata spiders were found on webs attached to the undersides of understory tree leaves. Waldorf (1976) found that E. ovata inhabited leaflet microhabitats on Mahonia aquifolium, a forest understory plant. Mature specimens of this spider species are found on bushes and trees in July, when mating occurs (Kaston 1948). Thus, growing spiders disperse from forest litter in summer and occupy microhabitats on leaves in the forest understory. These results are consistent with previously-documented evidence for spider growth and dispersal. Growth of forest litter spiders is often quite rapid in spring and early summer (Moulder and Reichle 1972). Associated with increased spider body size are increased needs for energy (food intake) and space (e.g., for larger webs). Dabrowska-Prot and Luczak (1968) and Enders (1974) demonstrate that, as web-building spiders mature and grow in size, they often choose web sites at greater heights above the ground than smaller conspecifics. These is also considerable evidence that larger spiders with webs well above the ground feed on larger prey (Enders 1974, Waldorf 1976). CONCLUSION Leaf shape is one variable in a complex association of variables which together consti- tute the forest litter habitat. Other such variables are litter depth and microclimate, which are easily measured, and number of microhabitats and spatial arrangements of litter units, which are less easily quantified. These latter variables may differ for individual species depending on spider size and spatial requirements. Leaf shape is shown to influence directly the number of species within forest litter and microhabitat selection of some species (e.g., E. ovata ) and higher taxa (e.g., Anyphaenidae). Future research should attempt to clarify whether distribution patterns of individual species are due to architec- tural properties of the substrate, to interspecific interactions such as predation, or to combinations of these and other factors. ACKNOWLEDGMENTS We wish to thank Myron J. Mitchell, Roy A. Norton and Neil H. Ringler for critically reviewing the manuscript. Susan E. Riechert and George W. Uetz provided useful com- ments on an earlier draft of the paper. This research was supported by funds from the Mclntire-Stennis Cooperative Forestry Program, USDA (SUNY RF Nos. 210-L007-H and I). LITERATURE CITED Bell, D. T. and S. K. Sipp. 1975. The litter stratum in the streamside forest ecosystem. Oikos, 26:391-397. STEVENSON AND DINDAL-LEAF SHAPE AND FOREST LITTER SPIDERS 177 Berry, J. W. 1967. The distributional ecology of spiders in the old field succession of the Piedmont region of North Carolina. Ph.D. Thesis, Duke University, Durham, North Carolina. Dabrow ska-Pro t, E. and J. Luczak. 1968. Spiders and mosquitoes of the ecotone of an alder forest ( Cared elongate-aluetum) and an oak-pine forest ( Pin o-Querce turn ). Ekol. Pol., 16 Ser A: 461-483. Dindal, D. L. and L. Levitan. 1977. The soil invertebrate community of composting toilet systems. Ecol. Bull. (Stockholm), 25:577-580. Duffey, E. 1966 Spider ecology and habitat structure. Senk. Biol., 47:45-49. Edgar, W. D. 1969. Prey and predators of the wolf spider, Lycosa lugubris. J. Zool., London, 159:405-411. Enders, F. 1973. Selection of habitat by the spider, Argiope aurantia Lucas (Araneidae). Amer. Midi. Nat., 90:47-55. Enders, F. 1974. Vertical stratification in orb-weavers (Araneidae, Araneae) and a consideration of other methods of coexistence. Ecology, 55:317-328. Greenquist, E. A. and J. S. Rovner. 1976. Lycosid spiders on artificial foliage: stratum choice, orientation preferences and prey wrapping. Psyche, 83:196-209. Hagstrum, D. W. 1970. Ecological energetics of the spider Tarantula kochi. Ann. Entomol. Soc. Am., 63:1297-1304. Hallander, H. 1970. Prey, cannibalism and microhabitat selection in the wolf spiders Pardosa chelata O. F. Muler andP. pullata Clerck. Oikos, 21:337-340. Jocque, R. 1973. The spider fauna of adjacent woodland areas with different humus types. Biol. Jaarboek, 41:153-179. Kaston, B. J. 1948. Spiders of Connecticut. State Geol. Nat. Hist. Surv. Conn. Bull. No. 70, 874 pp. Kaston, B. J. 1978. How to know the spiders, 3rd ed. William C. Brown Company, Publishers, Dubuque, Iowa. 272 pp. Lowrie, D. C. 1942. The ecology of spiders of the xeric dunelands in the Chicago area. Chicago Acad. Sci. Bull., 6:161-189. Lowrie, D. C. 1948. The ecological succession of spiders of the Chicago area dunes. Ecology, 29:334-351. Lubin, Y. D. 1978. Seasonal abundance and diversity of web-building spiders in relation to habitat structure on Barro Colorado Island, Panama. J. Arachnol., 6:31-51. MacArthur, R. H. and J. W. MacArthur. 1961. On bird species diversity. Ecology, 42:594-598. Moulder, B. C. and D. C. Reichle. 1972. Significance of spider predation in the energy dynamics of forest floor arthropod communities. Ecol. Monogr., 42:473-498. Pielou, E. C. 1966. The measurement of diversity in different types of biological collections. J. Theor. Biol., 13:131-144. Pritchard, R. P. 1941. Report on experimental planting, Syracuse forest experiment station. Bull. N. Y. State Coll. For. Tech. Publ., 57:1-43. Reice, S. R. 1980. The role of substratum in benthic macroinvertebrate microdistribution and litter decomposition in a woodland stream. Ecology, 61:580-590. Robinson, J. V. 1981. The effect of architectural variation in habitat on a spider community: an experimental field study. Ecology, 62:73-80. Sokal, R. R. and F. J. Rohlf. 1969. Biometry. W. H. Freeman, San Francisco. 776 pp. Stevenson, B. G. 1980. Effect of leaf shape on microhabitat selection and community structure of selected forest litter arthropods. M. S. Thesis, State University of New York, Syracuse, New York. Stevenson, B. G. and D. L. Dindal. 1981. A litter box method for the study of litter arthropods. J. Georgia Entomol. Soc., 16:151-156. Stratton, G. E., G. W. Uetz, and D. G. Dillery. 1979. A comparison of the spiders of three coniferous tree species. J. Arachnol., 6:219-226. Turnbull, A. L. 1960. The spider populations of a stand of oak ( Quercus robor L.) in Wytham Woods, Berks., England. Canad. Entomol., 92:110-124. Uetz, G. W. 1974. A method for measuring habitat space in studies of hardwood forest litter arthro- pods. Environ. Entomol., 3:313-315. Uetz, G. W. 1975. Temporal and spatial variation in species diversity of wandering spiders (Araneae) in deciduous forest litter. Environ. Entomol., 4:719-724. Uetz, G. W. 1976. Gradient analysis of spider communities in a streamside forest. Oecologia, 22:373-385. Uetz, G. W. 1979. The influence of variation in litter habitats on spider communities. Oecologia, 40:29-42. 178 THE JOURNAL OF ARACHNOLOGY USD A. 1972 Soil Series of the U.S., Puerto Rico and the Virgin Islands: Their Taxonomic Classifica- tion, Soil Cons. Serv. USGPO, Wash. D.C. Waldorf, E. S. 1976. Spider size, microhabitat selection and use of food. Amer. Midi. Nat., 96:76-87. Manuscript received March 1981, revised July 1981. Forester, L. 1982. Non-visual prey-capture in Trite planiceps, a jumping spider (Araneae, Salticidae). J. Arachnol., 10:179-183. NON-VISUAL PREY-CAPTURE IN TRITE PLANICEPS, A JUMPING SPIDER (ARANEAE, SALTICIDAE)1 Lyn M. Forster Otago Museum Gt. King St. Dunedin, New Zealand ABSTRACT Jumping spiders, well known for their visually-mediated reactions, employ a distinctive set of responses when catching prey in daylight. It is shown here, however, that the jumping spider Trite planiceps readily seizes live house flies in the dark, and that such behaviour is probably accomplished by means of vibratory signals. INTRODUCTION Of all the families of spiders, the salticids or jumping spiders alone have achieved distinction by virtue of a unique and remarkable set of eyes (see review by Forster, in press, a) which they use to find their way about, escape from enemies, hunt and catch prey (Homann 1928, Crane 1949, Drees 1952, Gardner 1964, 1966, Forster, 1977a, b, 1979b), interact with conspecifics and court prospective mates (Crane 1949, Jackson, in press; Forster, in press, a). The visual reactions of jumping spiders to prey consist of a series of events broadly categorized as Orientation, Pursuit and Capture (Forster 1977a) and mediated by differ- ent pairs of eyes (Homann 1928, Drees 1952, Land 1971, Forster 1979a). Further evi- dence that these spiders catch prey visually and hence only in daylight was provided by the ‘red light’ experiments of Jackson (1977) in which 33 adult Phidippus johnsoni were offered house flies in a situation where visible wavelengths shorter than 600 nm were excluded. None of the spiders caught flies despite their being within visual range nor were they seized upon contact, yet 15 min. later in white light 36% of them succeeded. The present study shows, however, that when Trite planiceps Simon 1899, are offered house flies in the dark they are indeed able to seize them and that they probably do so by means of vibratory cues. MATERIAL AND METHODS Sixteen adult Trite planiceps spiders (mostly female) were collected from flax (Phor- mium tenax ) in Taieri Beach Road, Dunedin, New Zealand. They were housed and 1 This study was supported by grants from the Scientific Distribution committee, Golden Kiwi Lottery Fund, to R. R. Forster. 180 THE JOURNAL OF ARACHNOLOGY maintained as described for previous studies (Forster 1977a, b). House flies were provided once a week and the present tests were conducted on a day on which they were normally fed. Sixteen test containers were prepared as follows: a small hole (1 cm diam.) was cut in the lid of a clear plastic petri dish (9 cm diam.). A hollow opaque plastic stopper containing a chilled fly was taped over this hole and the opening closed with a strip of cardboard (1.5 x 5 cm). A spider was then placed in the petri dish and covered with the prepared lid (Fig. 1). Tests were conducted in a photographic darkroom and containers were held here for 5 min. prior to testing to allow spiders to settle down after handling and flies to recover from refrigeration. Flies were released into the lower compartment by withdrawing the cardboard strip, this being quite easily accomplished in the dark. Fifteen minutes later spiders were examined in safelight conditions (to minimise the chances of captures occurring as the lights went on). In the second part of the investigation, 5 adult T. planiceps spiders (all female) were blinded by completely covering their eyes with black acrylic water-based paint (for details of method see Forster 1979a). They were placed in petri dishes overnight to allow full recovery from anaesthesia, and next day were offered live flies in these containers under subdued illumination to prevent flies from being too active. As controls, freshly freeze-killed (1) intact and (2) squashed flies were offered to blinded and non-blinded spiders. RESULTS During the 15 min. dark period after flies were released into the spider compartments, a succession of ‘buzzes’ lasting for about 20 sec. was clearly audible. At the end of the test period, inspection revealed that 14 of the 16 spiders (88%) were already consuming their partially liquified victims. There seems little doubt that the ‘buzzing’ sounds indi- cated the seizure of flies by spiders. Because this result was unexpected however, the test was repeated, this time with an 80% capture rate. Totally blinded spiders moved about very little in their containers even after 24 hr. but when they did, they walked slowly and hesitantly with much foreleg tapping. Upon the introduction of flies however, their activity increased quite markedly. They moved much Fig. 1.— Section through a test container showing fly in upper chamber separated by strip of cardboard from spider in lower compartment. The tape (not shown) holding the upper chamber in place is at right angles to the cardboard strip which can then be slid gently out from beneath the fly, allowing it to drop into the compartment below. FORSTER-NON-VISUAL PREY-CAPTURE IN A JUMPING SPIDER 181 faster, turning frequently but haphazardly in various directions. At the same time foreleg tapping changed to foreleg waving with these legs often being raised above their heads, sometimes in unison, at other times alternately. Flies colliding with the rear or lateral regions of the spider’s body provoked with- drawal but not escape or turning responses as would have been the case in the spider’s normal non-blinded state. However, flies which ran between the spider’s forelegs or which were held there with forceps were immediately seized. At once flies began struggling, and emitted high frequency wing vibrations, activities which usually ceased after 20-30 sec., presumably when the venom took effect. Blinded spiders did not accept freshly killed (1) intact or (2) squashed flies as prey when held with forceps, presented between their forelegs and brought into contact with the chelicerae. If such flies were left overnight in containers with spiders, whether blinded or non-blinded, intact flies were never accepted but squashed flies were generally eaten, in one test by 79% of non-blinded spiders. The consumption of flies by T. planiceps is easily verified because either their skeletal remains are discarded as a crushed ball or they are dismembered, with legs and wings then becoming scattered about the container. DISCUSSION Intact flies were seized by T. planiceps spiders in the dark, or when blinded, only when they were alive, suggesting that captures are not induced by chemical or tactile cues but by some other property of the living prey. Evidently squashed flies which were eaten differed in this respect; presumably spiders recognized the chemical substances emanating from extruded tissues and fluids, a proposal supported by earlier studies (see Kaestner 1968) which suggested that salticids can detect the presence of crushed flies beneath moist filter paper. ‘Dark’ conditions precluded the use of vision in the capture of prey and blinded spiders exhibited none of the usual reactions performed by spiders when catching flies in daylight. The most probable sensory modality involved in such captures is that of vibra- tion, clearly effective only at very close range. Active flies running between the forelegs of a hungry spider probably generate air-borne vibratory signals which evoke a series of motor actions resulting in prey seizure. Of course we should not discount the possibility that such a mechanism also operates in the final stages of visual prey-capture. Foreleg-waving by blinded spiders in the presence of active flies further supports the view that air-borne vibration is being generated by flies and detected by them, but at distances greater than 1 to 2 cm spiders are unable to determine the direction of such signals nor do they make any attempt to close the gap between them and the source of such signals. Crane (1949) proposed that many salticids make use of vibration in courtship while Forster (1979b and in press, a) showed, by transition-frequency analysis, that foreleg- waving in Trite auricoma plays little part in visual communication during courtship dis- plays. One possibility advanced in those investigations was that this posture serves to pick up non-courtship, environmental information such as vibration. The present observations add weight to such a role for foreleg- waving while studies of T. auricoma spiderlings show that these same movements increase in frequency and intensity when their prey (winged Drosophila ) are out of sight (Forster 1977a). One might well ask why such visually competent animals possess a means of catching prey in the dark, particularly when jumping spiders in general are assumed to rest at 182 THE JOURNAL OF ARACHNOLOGY night. It is not known of course whether other salticid species can also catch prey under such conditions but T. planiceps are unusual in that they do not make nightly web shelters. This means that without the barrier afforded by a silk retreat, nocturnal insects may run into them and thus be captured. Such an event, however, would merely supple- ment the usual quota caught in daylight. However, a more likely explanation relates to their survival in the cool overcast winters experienced at low altitudes in New Zealand. High altitude salticids, for example, spend the winter months under snow and have evolved mechanisms by which they cope with long periods without food (Forster, unpubl. obs.). Trite planiceps , on the other hand, spend much of the winter sheltering within the dim recesses of rolled-up flax leaves and need sustenance from time to time (Forster and Forster, 1973, 1976, Forster 1977b and in press, b). With a non-visual, energy-conserving tactic at their disposal they would be able to catch an occasional fly or moth at times when low light levels preclude the use of their visual response sequences. These findings also raise questions about the mode of prey-capture employed by those salticid species living 'in the dense, damp, evergreen forests of New Zealand and, more particularly, in the thick layers of leaf litter which lie beneath. Here, these spiders spend their lives in an environment buffered against the extremes of temperature and one into which the sunlight rarely penetrates. Can their reliance on vision in prey-capture (and courtship) be as great as in most of the jumping spiders we know, or do they depend more on vibration and perhaps even other sensory systems to carry out these basic tasks? ACKNOWLEDGMENTS I thank my husband, Ray Forster, for reading and commenting on the manuscript. REFERENCES Crane, J. 1949. Comparative biology of salticid spiders at Rancho Grande, Venezuela. Part IV: An analysis of display. Zoologica, 34:159-214. Drees, O. 1952. Untersuchunger iiber die angeborenen Verhaltensweisen bei Springspinnen (Salti- cidae). Z. Tierpsychol., 9:169-207. Forster, L. M. 1977a. A qualitative analysis of hunting behaviour in jumping spiders (Araneae: Salti- cidae). New Zealand Journ. Zool., 4:51-62. Forster, L. M. 1977b. Some factors affecting feeding behaviour in young Trite auricoma spiderlings (Araneae: Salticidae). New Zealand Journ. Zool., 4(4):435-443. Forster, L. M. 1979a. Visual mechanisms of hunting behaviour in Trite planiceps - a jumping spider (Araneae: Salticidae). New Zealand Journ. Zool., 6:79-93. Forster, L. M. 1979b. Comparative aspects of the behavioural biology of New Zealand jumping spiders (Araneae: Salticidae). Ph. D. dissertation, University of Otago, New Zealand. 402 pp. Forster, L. M. in press a. Visual communication in jumping spiders (Salticidae), chapter 5, in Spider Communication: Mechanisms and Ecological Significance. P. N. Witt and J. S. Rovner, eds. Prince- ton University Press. Forster, L. M. in press b. Vision and prey-catching strategies in jumping spiders. American Scientist. Forster, R. R. and Forster, L. J. 1973. New Zealand Spiders - an introduction. 254 pp. Collins, Auckland. London. Forster, R. R. and Forster, L. M. 1976. Jumping spiders, in New Zealand’s Heritage. R. Knox, ed. 6(76): 21 14-21 17 . Gardner, B. T. 1964. Hunger and sequential responses in the hunting behavior of salticid spiders. J. Comp. Physiol. Psychol., 59(2): 167-173. FORSTER- NON-VISUAL PREY-CAPTURE IN A JUMPING SPIDER 183 Gardner, B. T. 1966. Hunger and characteristics of the prey in the hunting behaviour of salticid spiders. J. Comp. Physiol. Psychol., 62(3):475-478. Homann, H. 1928. Bietrage zur Physiologie der Spinnen augen. Z. Vergl. Physiol., 7:201-268. Jackson, R. R. 1977. Courtship versatility in the jumping spider, Phidippus johnsoni (Araneae: Salti- cidae). Anim. Behav., 25:95 3-957. Jackson, R. R. in press. The behaviour of communication in jumping spiders (Salticidae). Chapter 6 in Spider Communication: Mechanisms and Ecological Significance. P. N. Witt and J. S. Rovner, eds. Princeton University Press. Kaestner, A. 1968. Invertebrate Zoology. Vol. 2:131-203. Translated by H. W. Levi and Lorna Levi. John Wiley and Sons, New York, London, Sydney. Land, M. F. 1971. Orientation by jumping spiders in the absence of visual feedback. J. Exp. Biol., 54:119-139. Manuscript received 9 April 1981, revised 15 July 1981 Opell, B. D. 1982. Post-hatching development and web production of Hyptiotes cavatus (Hentz) (Araneae, Uloboridae). J. Arachnol., 10:185-191. POST-HATCHING DEVELOPMENT AND WEB PRODUCTION OF HYPTIOTES CA VATUS (HENTZ) (ARANEAE, ULOBORIDAE) Brent D. Opell Department of Biology, Virginia Polytechnic Institute and State University, Blacksburg, Virginia 24061 ABSTRACT Members of Hyptiotes cavatus construct vertical triangle-webs, consisting of only four radii span- ned by capture threads. Spiderlings began constructing triangle capture webs only as third instars. In the laboratory male and female development was synchronous, most individuals maturing as sixth instars, at which time males ceased constructing capture webs. Capture webs were usually constructed only during the first sixty percent of a developmental stadium. Males and females reared in large containers produced more webs during and spent more time in the fifth stadium than those reared in small containers with more closely spaced supports. Development and web production of earlier instars were not significantly influenced by container size. INTRODUCTION Hyptiotes cavatus (Hentz) are common in the eastern half of the United States (Muma and Gertsch 1964) where their webs are often found in the lower branches of conifers. This relatively stable habitat affords a large number of web attachment points and seems to support a relatively constant population throughout the spring and summer. These spiders produce vertical triangle-webs which have only four radii with cribellar capture “spirals” extending between them (Fig. 1 ; Comstock 1913, Emerton 1902, Gertsch 1979, McCook 1884, Wilder 1875). Webs of this type are also produced by H. flavidus (Black- wall) (Castelnau and Thorell 1 897), H. gertschi Chamberlin and Ivie (Muma and Gertsch 1964; personal observations), and H. paradoxus (C. L. Koch) (Marples and Marples 1937, Peters 1938, Wiehle 1927) and are probably characteristic of the genus’ six other species. This web form is thought to be a reduced and reoriented horizontal orb-web, the primi- tive and most common web form of the family Uloboridae to which Hyptiotes belongs (Opell 1979). Associated with web reduction and change in orientation is a more active mode of web monitoring and use during prey capture. Rather than hanging beneath a horizontal web’s hub as the family’s orb-weaving members do, Hyptiotes hold their webs taut while resting near the attachment of the apex line (Fig. 1). Here their robust body form renders them inconspicuously bump-like. When a prey contacts the web the spider releases about a centimeter of slack silk held by the fourth legs, causing the web to be shaken, apparently to ensure prey entanglement and perhaps to allow evaluation of prey weight. After one or 186 THE JOURNAL OF ARACHNOLOGY several web jerks the spider runs down the apex line (often collapsing the web) and wraps the prey, initially by throwing silk over it and eventually by more complete wrapping as the prey is rotated while being held with the second and third legs. In the process of capturing even a small prey the web is often completely destroyed, eaten by the spider, and usually replaced the following night. This is in contrast with orb-weaving uloborids which often repair their webs before replacing them (Eberhard 1972; personal observa- tions). The life cycle of H. cavatus also contrasts with that of the eastern temperate orb- weaving species, Uloborus glomosus (Walckenaer). The former matures in late August and early September and females produce attached, plano-convex eggsacs, each containing five to 12 eggs, which pass the winter and from which spiderlings emerge early the following spring (Kaston 1948, Scheffer 1905, Wilder 1875; personal observations). Mem- bers of U. glomosus mature in June and early July and produce suspended, stellate eggsacs, each containing 30 - 60 eggs (Kaston 1948; personal observations). Spiderlings emerge from these eggsacs later the same summer and overwinter as immatures. Many questions about the biology of H. cavatus remain unanswered. The purpose of this study is to investigate the species’ post-hatching development and web construction. Specifically, this study will determine: 1) the number and duration of post-hatching stadia, 2) how many webs are produced during each stadium, 3) what portion of each stadium is devoted to prey capture, and 4) what effect sex and available support spacing have upon these factors. This quantitative information about the species’ development should be useful in designing future ethological and ecological studies of this intriguing species. Fig. l.-Webs of Hyptiotes cavatus adult female (foreground) and fifth instar male (background), 0.78 natural size. OPELL -HYPTIOTES CA VATUS DEVELOPMENT 187 METHODS AND MATERIALS Laboratory observations were conducted on spiders reared from eggsacs collected on 25 November 1978 from the lower limbs of a single hemlock ( Tsuga canadensis) near Newport, Giles County, Virginia. Eggsacs were refrigerated and, as spiders were needed, incubated at 23 to 25°C. Individuals were reared in separate plastic containers measuring either 30 x 16 x 8.5 cm or 34.5 x 25.5 x 16.5 cm. Wooden dowel rods cemented into larger containers formed a 30 x 23.5 cm rectangle at the container’s center. Those cemented into smaller containers formed polygons of various sizes, the smallest being a 14.5 x 13 x 6.5 cm triangle and the largest a 26.5 x 15.5 cm rectangle. All containers were kept in a chamber where conditions of 23 to 25°C, 85 to 95% relative humidity, and a 10:14 hour light: dark cycle were maintained. Laboratory studies were conducted from 14 January to 1 July 1979, during which time a total of eight males and 13 females were reared either to maturity (17 specimens) or to the penultimate instar, at which time the individual’s sex could be determined. Spiders were observed daily except for eight instances involving an average of seven specimens in which one day was missed, three instances involving an average of four specimens in which two days were missed, and one instance involving three specimens in which three days were missed. The feeding regime consisted of blowing one wild type Drosophila melanogaster into each web produced. Exlivia and alcohol preserved adults were placed in individual vials and carapace and femur measurements later taken at 50x with a micrometer-equipped dissecting microscope. I 1 1 II » II a * M g M 1 f_| 3 rd 4 th 5 th 6 th 7 th Fig. 2. -Changes in carapace (C) and fourth femur (F) lengths during development of male and female Hytiotes cavatus. 188 THE JOURNAL OF ARACHNOLOGY RESULTS Overview of Developmen t.-Hyptiotes cavatus eggsacs are small, dark brown to gray, oval, plano-convex objects that are tightly appressed to the small lower branches of conifers (Scheffer 1905). Each of 1 1 field-produced eggsacs opened contained seven to 12 (X = 10) 1 mm-diameter eggs. Spiderlings hatched and molted once within the eggsac before making a single, small exit hole in the eggsac’s convex surface. Like other ulo- borids, newly emerged second instars lacked a functional cribellum and calamistrum (structures used to make cribellar prey-capture silk which is stretched across the web’s four radii) and they hung for three to six days in an adult-like posture from short horizontal or diagonal threads until they molted a second time. Possessing a functional cribellum and calamistrum, these third instars constructed capture webs typical of the species. Adult male H. cavatus lacked a functional cribellum and calamistrum and did not construct capture webs. Males lived from 16 to 24 days after reaching maturity. Third instar spiderlings constructed webs in the small container’s smallest polygons and appeared to have difficulty producing webs in the large container. For this reason, most spiderlings were transfered to large containers only after they became fourth instars. Third instars successfully subdued, wrapped, and fed upon the same strain of D. melanogaster given subsequent instars. Two males and two females matured as seventh instars and had longer carapaces and fourth femora than the remaining six males and 1 1 females which matured as sixth instars (Fig. 2). This is similar to Berland’s (1914) findings that Uloborus plumipes Lucas ma- tured in five molts. Within a given stadium abdominal enlargement was evident and similar in males and females. After molting, these gains were negated and the abdomen assumed the same proportionally slender appearance seen early in the previous stadium. Carapace and femur lengths of females usually exceeded those of males in the same stadia (Fig. 2), but T-tests showed these differences to be significant (p < 0.05) only for sixth instar carapace length. Duration of Stadia.— The duration of each stadium and the proportion of this time during which webs were constructed are given in Figure 3 and the number of webs produced in Table 1. Webs were not constructed during the latter 35-50% of each devel- opmental stadium. This period (REST) was subtracted from the length of each stadium (DURATION) to determine the number of potential feeding days (WEB-CONSTRUC- TION). DURATION represents total time devoted to a developmental stage, WEB- CONSTRUCTION; the time spent acquiring energy and material necessary for growth; and REST, the final phase of development signifying either that enough resources have been obtained and/or that a phase has been entered in which prey capture activities are either not possible or would interfere with developmental events. The independent and crossed effects of container size and sex upon DURATION, WEB-CONSTRUCTION, and REST in the third, fourth, and fifth stadia and upon the total days spent in each period were tested using Analysis of Variance tests (ANOVA). The only parameters significantly influenced (p < 0.05) were DURATION and WEB- -CONSTRUCTION periods of the fifth stadium, both of which were affected by con- tainer size. In small containers males spent an average of 16.0 days in the fifth stadium, 10.7 of those being feeding days. Female values were 17.1 and 1 1 .1 , respectively. In large containers values for males were 18.0 and 12.0, respectively and those for females 21.5 and 16.2, respectively. There were no significant differences between the DURATION, WEB-CONSTRUC- TION, and REST periods of the third, fourth, and fifth stadia, as tested with ANOVA. 0?ELL-HYPTIOTES CA VATUS DEVELOPMENT 189 Number of Webs.— Independent and crossed effects of container size and sex upon the number of webs produced during third, fourth, and fifth stadia and upon the total number of webs produced during these three developmental stadia were tested with ANOVA. Of these parameters, only instar significantly influenced (p < 0.05) the number of webs produced. The greatest number of webs were produced during the fifth stadium and the least during the fourth stadium (Table 1). DISCUSSION Overwintering by Hyptiotes cavatus differs from that of H. paradoxus. During the summer, populations of the latter species are comprised both of individuals which have emerged from eggsacs early in the spring and half-grown individuals which emerged late the previous summer and passed the winter as immatures (Wiehle 1927). In H. cavatus DAYS Fig. 3. -Developmental periods of Hyptiotes cavatus third, fourth, fifth, and sixth instar males and females reared in small and large containers. Intervals at the right of boxes represent sd of stadia and those above sd of web construction periods. 190 THE JOURNAL OF ARACHNOLOGY Table l.-Web production during Hyptiotes cavatus development [X and (sd) ] INSTAR SMALL CONT. LARGE CONT. BOTH CONTAINERS male female male female male female all Third 4. 6(2. 2) 4.7(2. 3) __ 4.0 4. 6(2. 2) 4. 6(2. 2) 4.6(2. 1) Fourth 3.5(1. 1) 3. 3(0.7) 2.0 3. 7(1. 2) 3. 1(1.1) 3.4(0. 8) 3. 3(0.9) Fifth 4. 7(1. 4) 4. 9(1. 6) 5. 5(0. 7) 5. 5(1. 4) 4.9(1. 3) 5. 2(1. 5) 5. 1(1.4) TOTAL 13.3(3.4) 13.4(3.1) — 14.0 13.3(3.5) 13.5(2.8) 13.5(3.0) SAMPLE SIZE 6 7 2 6 8 13 21 only third instar individuals and webs were seen in the spring of the two consecutive years. Likewise, adults were seen only in late summer. This discrepancy may represent a population rather than a species difference, but published accounts of H. cavatus repro- duction and life history are not specific enought to resolve the question. Unlike members of several orb-weaving uloborid genera (Eberhard 1969, 1977, Opell 1979, Peters 1953, Szlep 1961, Wiehle 1927), second instar Hyptiotes did not construct a non-cribellar horizontal mesh web for prey capture. Instead, they hung from resting threads for several days until they molted again and were able to construct triangle-webs. Inactivity of second instars is similar to that reported for Miagrammopes species (Lubin et al. 1978), but is of longer duration. Because the primitive uloborid genera (Opell 1979) Tangaroa and Waitkera construct orb-webs (J. A. Beatty, personal communication, Forster 1967), the orb-web must be considered plesiomorphic for the Uloboridae and the webs of Hyptiotes and Miagrammopes apomorphic. Structural and behavioral specializa- tions associated with construction of these two reduced (specialized) web forms may preclude production of a non-cribellar capture web similar to that made by young orb- weavers and favor a short, non-feeding second stadium. Third instar spiderlings were collected in the field as early as 7 July and adults as early as 23 August, indicating that mean laboratory developmental times of 55-60 days (for spiders which mature in the sixth stadium) were close to field development times. Be- cause lab-reared specimens were fed only a small prey item each time they made a web and because prey capture usually resulted in web destruction, laboratory results may overestimate the number of webs normally produced. However, if spiders normally cap- ture less than one prey per day these results more nearly reflect field averages. Under the laboratory feeding regime male and female development was synchronous. DURATION, WEB-CONSTRUCTION, and REST periods of third and fourth stadia were statistically constant in large and small containers. However, in the fifth stadium spiders reared in large containers had longer DURATIONS and WEB-CONSTRUCTION periods. As nearly all web and wrapping silk was reingested by the spiders, this effect was probably due to the greater amount of energy required to produce larger webs found in large containers. This explanation is supported by a higher mean number of webs pro- duced by fifth instars of both sexes reared in large containers (Table I). Since lab-reared spiders were fed one fly per web and very few webs were destroyed while being measured (most damaged webs were third instar webs), the number of webs is a good index of energy requirements during fourth and fifth stadia. As shown by time spent feeding on Drosophila and by dry weights of discarded prey, fourth, and fifth, and sixth instars extract a fairly constant amount of material from each prey (Opell, in preparation). This indicates that in order to construct their webs in large containers fifth instar females OPELL -HYPTIOTES CA VATUS DEVELOPMENT 191 required additional energy roughly equivalent to that derived from 0.6 Drosophila. This is in general agreement with Peakall and Witt’s (1976) finding that silk formation and web construction require relatively little energy. Ethological and ecological contrasts between H. cavatus and other uloborids as well as orb-weavers of other families make this species an excellent subject for future study. The rearing procedures, developmental observations, and influence of support spacing upon development presented in this study should facilitate such studies. ACKNOWLEDGMENTS I thank Robin M. Andrews for making comments on an earlier version of this manu- script. LITERATURE CITED Berland, J. 1914. Note sur le cycle vital d’une Araignee cribellate, Uloborus plumipes Lucus. Arch. Zool. Exper., 54 N R:45-57. Castelnau, J. and T. Thorell. 1897. Notes sur Hyptiotes anceps. Feuille J. Nat., (3)27:107-111. Comstock, J. H. 1913. The Spider Book. Garden City, New York: Doubleday, Page, and Co. Eberhard, W. G. 1969. The spider Uloborus diversus Marx (Uloboridae) and its web. Ph.D. Thesis, Harvard University, Cambridge, Massachusetts. Eberhard, W. G. 1972. The web of Uloborus diversus (Araneae: Uloboridae). J. Zool. London, 166:417-465. Eberhard, W. G. 1977. The webs of newly emerged Uloborus diversus and of the male Uloborus sp. (Araneae, Uloboridae). J. Arachnol., 4:201-206. Emerton, J. H. 1902. The Common Spiders of the United States. Boston: Ginn and Co. Forster, R. R. 1967. The spiders of New Zealand (part 1). Otago Mus. Bull., 3:1-184. Gertsch, W. J. 1979. American Spiders, 2nd ed. New York: Van Nostrand Reinhold Co. Kaston, B. J. 1948. Spiders of Connecticut. Conn. Geol. Nat. Hist. Surv. Bull., 70:1-874. Lubin, Y. D., W. G. Eberhard, G. G. Montgomery. 1978. Webs of Miagrammopes (Araneae: Ulo- boridae) in the Neotropics. Psyche, 85(1): 1-23. McCook, H. C. 1894. American Spiders and their Spinning-work. Vol. 3. Philadelphia, pp. 1-285. Marples, M. J. and B. J. Marples. 1937. Notes on the spiders Hyptiotes paradoxus and Cyclosa conica. Proc. Zool. Soc. London (A), 107:213-221. Muma, M. H. and W. J. Gertsch. 1964. The spider family Uloboridae in North America north of Mexico. Amer. Mus. Novitates, 2196:1-43. Opell, B. D. 1979. Revision of the Genera and Tropical American Species of the Spider Family Uloboridae. Bull. Museum Comparative Zoology, 148(10):443-549. Peakall, D. B. and P. N. Witt. 1976. The energy budget of an orb web-building spider. Comp. Biochem. Physiol., 54 A (2):187-190. Peters, H. M. 1938. Uber das Netz der Dreieckspinne, Hyptiotes paradoxus. Zool. Anz. 121(3-4):49-59. Peters, H. M. 1953. Beitrage zur vergleichenden Ethologie und Okologie tropischer Webspinnen. Zeits. Morph. Okol. Tiere, 42:278-306. Scheffer, T. H. 1905. The cocooning habits of spiders. Kansas Univ. Sci. Bull., 3(2) :85-l 14. Szlep, R. 1961. Developmental changes in web-spinning instinct of Uloboridae: construction of the primary-type web. Behaviour, 27:60-70. Wiehle, H. 1927. Beitrage zur Kenntnis des Radnetzbaues der Epeiriden, Tetragnathiden und Ulo- boriden. Zeits. Morph. Okol. Tiere, 8(3-4) :468-5 37. Wilder, B. G. 1875. The Triangle Spider. Pop. Sci. Monthly, 1875(April):l-15. Manuscript received June 1 981, revised August 1 981 The Journal of Arachnology 10:192 RESEARCH NOTES A PARASITIC NEMATODE (MERMITHIDAE) FROM THE PSEUDOSCORPION “STERNOPHORUS” HIRSTI CHAMBERLIN (STERNOPHORIDAE) Parasitic nematodes of the family Mermithidae are not uncommonly found within in- vertebrates (de Coninck 1965, Traite de Zoologie 4:683-690). Kaston (1945, Trans. Conn. Acad. Arts Sci. 36:241-244) and Parker and Roberts (1973, Bull. Brit, arachnol. Soc. 3:82-84) have noted several occurrences within spiders. However, only one case has been recorded for pseudoscorpions (Vachon 1949, Traite de Zoologie 6:475), where six juvenile specimens of Hexamermis sp. were found within a female of Roncus sp. (Neobisiidae). This note reports the discovery of a juvenile mermithid nematode within a female of “Sternophorus” hirsti Chamberlin (Sternophoridae) (its generic position is soon to be changed, Harvey, in prep.), a common species which is widely distributed in eastern Australia. The sternophorid was alive at the time of collection, although the mermithid was not noticed until the host was cleared in lactic acid. Even though the parasite had not pierced through the cuticle of the pseudoscorpion, it occupied most of the abdominal cavity (Fig. 1) and its oral and caudal ends were juxtaposed against the ventral abdominal surface. The mermithid was dull orange in colour, the usual colour of the abdominal contents in the Sternophoridae. Fig. 1. -Mermithid nematode within abdomen of “ Sternophorus ” hirsti Chamberlin, dorsal view. Abbreviations: CA, carapace; Tl-Tll, terga 1 to 11. The nematode has been dissected out of its host and preserved separately. Specimen examined.— Female (MH302.51) of S. hirsti, with juvenile mermithid parasite, from under bark of Eucalpytus sp., 5 3 km W. of Tenterfield, New South Wales, Australia, 13 May 1981 (M. S. Harvey and M. Kotzman). Deposited in National Museum of Victoria. Mark S. Harvey, Department of Zoology, Monash University, Clayton, Victoria, 3168, Australia. Manuscript received July 1981, accepted September 1981. THE AMERICAN ARACHNOLOGICAL SOCIETY President : Jonathan Reiskind (1981-1983) Department of Zoology University of Florida Gainesville, Florida 32601 Membership Secretary: Norman I. Platnick (appointed) American Museum of Natural History Central Park West at 79th Street New York, New York 10024 Secretary: William A. Shear (1980-1982) Department of Biology Hampden-Sydney College Hampden-Sydney, Virginia 23943 President-Elect: Susan E. Riechert (1981-1983) Department of Zoology University of Tennessee Knoxville, Tennessee 37916 Treasurer: Norman V. Horner (1981-1983) Department of Biology Midwestern State University Wichita Falls, Texas 76308 Directors: Herbert W. Levi (1981-1983) Michael E. Robinson (1981-1983) Vincent D. Roth (1980-1982) The American Arachnological Society was founded in August, 1972, to promote the study of the Arachnida, to achieve closer cooperation between amateur and professional arachnologists, and to publish The Journal of Arachno logy. Membership in the Society is open to all persons interested in the Arachnida. Annual dues are $20.00 for regular members, $15.00 for student members. Correspondence con- cerning membership in the Society must be addressed to the Membership Secretary. Members of the Society receive a subscription to The Journal of Arachnology. In addi- tion, members receive the bi-annual newsletter of the Society, A merican Arachnology. American Arachnology , edited by the Secretary, contains arachnological news and comments, requests for specimens and hard-to-fmd literature, information about arach- nology courses and professional meetings, abstracts of the papers presented at the Society’s meetings, address changes and new listings of subscribers, and many other items intended to keep arachnologists informed about recent events and developments in arach- nology. Contributions for American Arachnology must be sent directly to the Secretary of the Society. The Eastern and Western sections of the Society hold regional meetings annually, and every three years the sections meet jointly at an International meeting. Information about meetings is published in American Arachnology, and details on attending the meetings are mailed by the host(s) of each particular meeting upon request from interested persons. CONTENTS THE JOURNAL OF ARACHNOLOGY VOLUME 10 SUMMER 1982 NUMBER 2 Feature Articles Pseudoscorpions of the family Chernetidae newly identified from Oregon (Pseudoscorpionida, Cheliferoidea), Ellen M. Benedict and David R. Malcolm 97 Predation by a commensal spider, Argyrodes trigonum , upon its host: an experimental study, David H. Wise Ill Natural history of a tropical, shrimp-eating spider (Pisauridae), Fredrica H. van Berkum 117 Solifugos de Colombia y Venezuela (Solifugae, Ammotrechidae), Emilio A. Maury 123 Las especies del genero Porrimosa Roewer, 1959 (Araneae, Hippasinae), Roberto M. Capocasale 145 Birth behavior in Diplocentrus bigbendensis Stahnke (Scorpiones, Diplocentridae), Oscar F. Francke 157 Effect of leaf shape on forest litter spiders: Community organization and microhabitat selection of immature Enoplognatha ovata (Clerck) (Theridiidae), Bruce G. Stevenson and Daniel L. Dindal 165 Non-visual prey-capture in Trite planiceps , a jumping spider (Araneae, Salticidae), Lyn Forster 179 Post-hatching development and web production of Hyptiotes cavatus (Hentz) (Araneae, Uloboridae), Brent D. Opell 185 Research Notes A parasitic nematode (Mermithidae) from the pseudoscorpion “Sternophorus” hirsti Chamberlin (Sternophoridae), Mark S. Harvey 192 Cover illustration, Alacran tartarus Francke, by Oscar F. Francke Printed by The Texas Tech University Press, Lubbock, Texas Posted at Crete, Nebraska, in September 1982 , f The Journal of ARACHNOLOGY OFFICIAL ORGAN OF THE AMERICAN ARACHNOLOGICAL SOCIETY VOLUME 10 FALL 1982 NUMBER 3 THE JOURNAL OF ARACHNOLOGY EDITOR : Oscar F. Francke, Texas Tech University. ASSOCIATE EDITOR : B. J. Kaston, San Diego State University. EDITORIAL ASSISTANT: James C. Cokendolpher, Texas Tech University. EDITORIAL BOARD: Charles D. Dondale, Agriculture Canada. William G. Eberhard, Universidad de Costa Rica. Maria E. Galiano, Museo Argentino de Ciencias Naturales. Willis J. Gertsch, American Museum of Natural History. Neil F. Hadley, Arizona State University. B. J. Kaston, San Diego State University. Herbert W. Levi, Harvard University. Emilio A. Maury, Museo Argentino de Ciencias Naturales. William B. Muchmore, University of Rochester. Martin H. Muma, Western New Mexico University. William B. Peck, Central Missouri State University. Norman I. Platnick, American Museum of Natural History. Susan E. Riechert, University of Tennessee. Michael E. Robinson, Smithsonian Tropical Research Inst. Jerome S. Rovner, Ohio University. William A. Shear, Hampden-Sydney College. Carlos E. Valerio, Universidad de Costa Rica. Stanley C. Williams, San Francisco State University. THE JOURNAL OF ARACHNOLOGY is published in Spring, Summer, and Fall by The American Arachnological Society in cooperation with The Graduate School, Texas Tech University. Individual subscriptions, which include membership in the Society, are $20.00 for regular members, $15.00 for student members. Institutional subscriptions to The Journal are $25.00. Correspondence concerning subscription and membership should be addres- sed to the Membership Secretary (see back inside cover). Back issues of The Journal are available from Dr. William B. Peck, Department of Biology, Central Missouri State Univer- sity, Warrensburg, Missouri 64093, U.S.A., at $5.00 for each number. Remittances should be made payable to The American Arachnological Society. Change of address notices must be sent to both the Secretary and the Membership Secretary. Manuscripts for THE JOURNAL OF ARACHNOLOGY are acceptable only from cur- rent members of the Society, and there are no page charges. Manuscripts must be type- written double or triple spaced on 8.5 in. by 11 in. bond paper with ample margins, and may be written in the following languages: English, French, Portuguese, and Spanish. Contributions dealing exclusively with any of the orders of Arachnida, excluding Acari, will be considered for publication. Papers of a comparative nature dealing with che Iter- ates in general, and directly relevant to the Arachnida are also acceptable. Detailed instructions for the preparation of manuscripts appear in the Fall issue of each year, and can also be obtained from the Editor and the Associate Editor. Manuscripts that do not follow those instructions will be returned to the author(s) without the benefit of review. Manuscripts and all related correspondence must be sent to Dr. B. J. Kaston, 5484 Hewlett Drive, San Diego, California 92115, U.S.A. Muchmore, W. B. 1982. The genera Ideobisium and Ideoblothrus, with remarks on the family Syarinidae (Pseudoscorpionida). J. Arachnol., 10:193-221. THE GENERA IDEOBISIUM AND IDEOBLOTHRUS , WITH REMARKS ON THE FAMILY SYARINIDAE (PSEUDOSCORPIONIDA) William B. Muchmore Department of Biology University of Rochester Rochester, New York 14627 ABSTRACT Ideobisium Balzan and Ideoblothrus Balzan are redefined after examination and redescription of the types, Ideobisium crassimanum Balzan and Ideoblothrus similis Balzan. Pachychitra Chamberlin is shown to be a synonym of Ideoblothrus. All known species of each genus are listed and doubtful species are discussed; Ideobisium balzanii With and Ideoblothrus seychellesensis (Chamberlin) are redescribed; and the following new species are described: Ideobisium chapmani, I. peckorum, I. ecuadorense, I. puertoricense, /. yunquense, and Ideoblothrus kochalkai, /. colombiae. The assignment of these genera to the family Syarinidae is discussed. INTRODUCTION The genus Ideobisium was created by Balzan (1891) with I. crassimanum , sp. nov., as its type species. At the same time, another species, I. similis, sp. nov., was also placed in the genus, but in a new subgenus named Ideoblothrus, primarily on the basis of its lack of eyes. [Also at that time, Balzan considered Ideoroncus Balzan 1890 as a subgenus of Ideobisium, but Ideoroncus has subsequently been recognized as a distinct and very different genus (Chamberlin 1930)] . Balzan placed Ideobisium in his new family Pseudo- bisiidae, which was differentiated from Obisiidae Hagen by the possession of a distinct process (galea) on the movable finger of the chelicera. Chamberlin (1930) rejected the name Pseudobisiidae because there is no genus Pseudobisium and substituted the name Ideobisiinae, with Ideobisium as type genus; he considered the group a subfamily of his newly erected family Neobisiidae, with the presence of a distinct galeal process as the distinguishing character. Beginning with Chamberlin (1930) the name Ideoblothrus has been ignored and species have been assigned to Ideobisium (without subgeneric distinc- tion) with no regard for the presence or absence of eyes. Several years ago, while studying examples of Ideobisium Balzan and Pachychitra Chamberlin from the Caribbean area, I noted the many similarities of the two genera. But it was only after the discovery of the lanceolate form of trichobothrium t (Muchmore 1979) that I was able to recognize the true nature of their relationship. Mahnert 194 THE JOURNAL OF ARACHNOLOGY (1979:750) also noted the resemblance of Ideobisium and Pachychitra, but was unable to carry his analysis to a satisfactory conclusion. As will be shown below, Pachychitra Chamberlin is actually a synonym of Ideo- blothrus Balzan, which is indeed closely related to Ideobisium Balzan, and both are referrable to the family Syarinidae. I have been able to study the types of /. crassimanum Balzan and I. similis Balzan through the kind cooperation of Dr. J. Heurtault, Museum National d’Histoire Naturelle, Paris. Also, I have seen a large amount of material pertaining to Ideobisium and Ideoblothrus from North, Middle and South America, the Caribbean area, and scattered other parts of the world. Depositories referred to in the text are abbreviated thus: [ACC] [AMNH] [BM(NH)] [FSCA] [JCC] [MCZ] [MNHN] Academia de Ciencias de Cuba, La Habana. American Museum of Natural History, New York, N. Y. British Museum (Natural History), London. Florida State Collection of Arthropods, Gainesville, Florida. (My own specimens are deposited here.) J. C. Chamberlin Collection, Forest Grove, Oregon. Museum of Comparative Zoology, Harvard University, Cambridge, Massachusetts. Museum National d’Histoire Naturelle, Paris. SYSTEMATICS Ideobisium Balzan Ideobisium Balzan 1891:543, Fig. 33, 33a; Chamberlin 1930:23, 36; Beier 1932:157; Mahnert 1979:750. Type species Ideobisium crassimanum Balzan 1891, by original designation. Diagnosis (revised).— A genus of the superfamily Neobisioidea Chamberlin (1930:7). Species generally small and robust. Carapace about square; anterior margin with a broad, low, rounded epistome; surface smooth; four distinct eyes; 22-28 vestitural setae, with four at anterior and six to eight at posterior margins. Apex of palpal coxa acute, with two long setae. Tergites and sternites entire, except sternites 2-5 often weakly divided; surfaces smooth; tergites 1-3 usually with six and tergite 4 with seven or eight setae; pleural membranes granulate anteriorly, becoming longitudinally granulo-striate or smoothly striate posteriorly. Internal genital setae of male arranged as two parallel, longitudinal rows of three. Cheliceral fingers toothed; hand with five acuminate setae, es shortest; flagellum of six or seven setae, all finely serrate except the basal one, which is also shorter than the others; galea long, simple. Palp robust, none of the segments more than 3.0 times as long as broad; all surfaces smooth; movable chelal finger shorter than the chelal hand; venom apparatus developed only in fixed finger, with duct short, extending less than half the distance to trichobothrium et\ chelal teeth contiguous, at least the distal ones cusped. Trichobothrium t on movable chelal finger shorter than all others and lanceolate at tip; t, st and sb usually closely grouped in an obliquely oriented series, with t proximad of middle of finger; on fixed finger it nearer level of et than of est; isb on external side of hand near base of fixed finger; eb and esb on side of hand, usually near the middle (Fig. 1). Leg I with basifemur 1.25 or more times as long as telofemur. Leg IV with femur distinctly indented dorsally at suture between its parts, the suture MUCHMORE-THE GENERA IDEOBISIUM AND IDEOBLOTHR US 195 itself slightly oblique to long axis of femur; tibia and each tarsal segment with a prominent tactile seta proximad of middle. Subterminal tarsal setae finely denticulate in distal half; arolia shorter than claws. Distribution.— Best known from northern South America and the West Indian islands, it is also represented in New Zealand and New Caledonia. Remarks.— Because of its overall similarity to Ideoblothrus , especially in the possession of only two setae on the apex of the palpal coxa and of a lanceolate trichobothrium t , and in spite of the granulate pleural membranes, Ideobisium must now be considered a member of the family Syarinidae, not of the Neobisiidae. With the removal of Ideobisium from the family Neobisiidae, the name Ideobisiinae is no longer valid for the group of neobisiids with distinct galeal processes. In view of the many uncertainties about relationships among the Neobisiidae, it seems best to refrain from naming subfamilies until more comparative work is done. Ideobisium crassimanum Balzan Figs. 2-9 Ideobisium crassimanum Balzan 1891:541; Chamberlin 1930:36-37; Beier 1932:158; nec Beier 1976:46. Material examined.— Holotype female from Caracas, Venezuela (Col. Mus. 14.155) [MNHM] . The specimen is mounted on two microscope slides, numbered 442 and 443; the left chelicera and the left palpal tibia and chela are missing, and the right palpal femur and chela are partially crushed. Also at hand are two males and three females from Rancho Grande, Aragua, Venezuela, 19-27 February 1971, S. Peck [FSCA] . Description.— Based on the holotype and the other five adults, considered conspecific in spite of some differences. Males and females similar in form, but females a little larger. Carapace and palps well sclerotized and reddish brown, other parts lighter brown. Carapace smooth; epistome broad, low, rounded (Fig. 2); four large corneate eyes; chaetotaxy of holotype 4-4-4-2-7 = 21, of others usually 4-4-4-6-6 = 24. Coxal area generally typical of the Neobisiodiea; apex of palpal coxa acute, bearing two long setae. Fig. 1 -Ideobisium sp.: lateral view of palpal chela, showing locations of trichobothria. 196 THE JOURNAL OF ARACHNOLOGY Abdomen elongate; tergites and posterior sternites entire. Anterior sternites with a tendency toward division, as evidenced by faint longitudinal wrinkles at the midline; surfaces smooth; pleural membranes heavily granulated anteriorly, becoming longitudi- nally striate posteriorly. Tergal chaetotaxy of holotype female 6:6:8:8:8:9:10: 9:9:8:T1T1T1T:2; sternal chaetotaxy of same 8:(3)1 0(3):(3)6(3): 12:13:1 1:11: 10:9 :T1T:2; anterior sternites of males with about 8: [3-3] :(3)5/7(3):(3)6(3):-; genital areas of female and male shown in Figs. 3 and 4. Chelicera about 0.6 as long as carapace; hand with five acuminate setae: flagellum of six or seven setae, equal in length except the proximal one shorter, all finely denticulate except the proximal one; fixed finger with about 12 small, and movable finger with about 10 larger, teeth; galea of both sexes long, slender, gently curved (Fig. 5). Palp stout (Fig. 6); femur 2.35-2.55, tibia 1.85-1.9, and chela (without pedicel) 2.05-2.1 times as long as broad; hand (without pedicel) 1.15-1.25 times as long as deep; movable finger 0.83-0.89 as long as hand. All surfaces smooth. Fixed chelal finger with 29-32 and movable finger with 37-42 cusped, marginal teeth. Trichobothria of chela as shown in Fig. 7 ; on movable finger, t is shorter than the others and lanceolate near the tip, as illustrated (Fig. 8); isb is on the side of the hand near base of fixed finger, while eb and esb are situated more proximally and ventrally. Legs rather short and stout. Leg I with basifemur about 1.3 times as long as telofemur; leg IV (Fig. 9) with entire femur 2.4-2.65 and tibia 3.9-4.15 times as long as deep. Dorsal contour of femur of leg IV broken by a distinct indentation at junction of basifemur and telofemur; the line of articulation between the two segments slightly oblique to the long axis of the femur as a whole. Subterminal tarsal setae finely pinnate along ventral sides; arolia shorter than claws. Tibia, metatarsus and telotarsus of leg IV each with a long tactile seta proximal to the middle. Measurements (mm).— Holotype female. Body length 1.75. Carapace length 0.64. Chelicera 0.39 by 0.185. Palpal femur 0.55 by 0.215; tibia 0.50 by ?; chela (without pedicel) 0.93 by ?; hand (without pedicel) 0.54 by ?; pedicel 0.075 long; movable finger 0.45 long. Leg I: basifemur 0.27 long; telofemur 0.20 long. Leg IV: entire femur 0.57 by 0.235; tibia 0.435 by 0.1 1 ; metatarsus 0.16 by 0.08; telotarsus 0.26 by 0.07. Ranges of five adults from Rancho Grande.— Body length 1.68-2.03. Carapace length 0.57-0.64. Chelicera 0.31-0.385 by 0.155-0.185. Palpal femur 0.445-0.52 by 0.19-0.22; tibia 0.41-0.49 by 0.215-0.25; chela (without pedicel) 0.695-0.88 by 0.33-0.415; hand (without pedicel) 0.41-0.495 by 0.33-0.42; pedicel 0.06-0.07 long; movable finger 0.355-0.43 long. Leg I: basifemur 0.22-0.26 long; telofemur 0.185-0.20 long. Leg IV: entire femur 0.48-0.555 by 0.19-0.22; tibia 0.37-0.435 by 0.095-0.105; metatarsus 0.14-0.16 by 0.075-0.08; telotarsus 0.215-0.27 by 0.06-0.065. Remarks.— Based on the measurements given by Balzan (1891:543) this species has always been characterized as having very short fingers on the palpal chela (Beier 1932:157). However, examination of the type specimen has shown that the fingers, while short compared with many pseudoscorpions, are about the same relative length as those in other Ideobisium species. Evidently some error was made in Balzan’s original report. As mentioned above, there is a difference in the chaetotaxies of the carapace between the holotype from Caracas and the specimens from Rancho Grande; in the row just anterior to the posterior marginal row, the holotype has two setae, while the others have six. Whether this difference is significant is not clear, because other characters are very similar. The problem can be resolved only by study of further material from the vicinity of Caracas. MUCHMORE-THE GENERA IDEOBISIUM AND IDEOBLOTHRUS 197 Figs. 2-9 .-Ideobisium crassimanum Balzan: 2, anterior margin of carapace; 3, genital opercula of female; 4, genital opercula (left) and internal genitalia (right) of male; 5, tip of movable finger of chelicera; 6, dorsal view of left palp; 7, lateral view of right chela; 8, trichobothrium t\ 9, leg IV. Three specimens from Trinidad may belong here, though they have some characters intermediate between this species and I. balzanii With (see below). Proper understanding of their position can come only after study of more material from eastern Venezuela and the Lesser Antilles. The specimens reported by Beier (1976) from the Dominican Republic certainly do not pertain to I. crassimanum , but rather to one of the species described below from Puerto Rico or to an undescribed indigenous species. Ideobisium balzanii With Fig. 10 Ideobisium Balzanii With, 1905:131. /. balzanii : Chamberlin 1930:37, 1931, Fig. 35,0; Hoff 1945:1. I. balzani : Beier 1932:158. 198 THE JOURNAL OF ARACHNOLOGY Material examined— Paratype male (JC 12.01001) from St. Vincent, West Indies [JCC] ; three males and three females from “tamisage de hojarosca” at “Morn a Louis” (= Morn a l’Eau?), Guadeloupe, French West Indies, 15 March 1975 (F. Chalumeau), [ACC] ; one male from Laudat, Dominica, West Indies, 12 June 1911, [AMNH] : one male from Long Ditton, Dominica, 21 June 1911, [AMNH] . Diagnosis— Generally similar to I. crassimanum, but with less robust palps, 1/w ratio of femur being 2. 6-2. 8 and that of chela 2.35-2.5. Description— The description given by With (1905:131-135 and Figs. 2a-h) for speci- mens from St. Vincent is generally accurate and fairly complete. Here are provided some supplemental data and measurements for the specimens listed above. The sexes generally similar but female a little larger than male. Chaetotaxy of carapace usually 4-4-4-4-6 = 22. Palpal coxa with two long setae at apex. Abdominal tergites and posterior sternites entire, anterior sternites often divided by faint sutures along the midline; pleural membranes heavily granulate anteriorly, becoming sparsely granulate posteriorly. Tergal chaetotaxy about 6:6:6:8:8:8:8:9:9:7:T1T1T1T:2; sternal chaeto- taxy of male about 8: [3-3] :(3)14(3):(3)6(3): 10: 12: 10: 10: 10:8:T1T:2. Genital opercula of female with 8-10 setae on face of each. Chelicera about 0.6 as long as carapace; hand with five acuminate setae; flagellum of six or seven setae, the proximal one shorter than the others. Palpal segments entirely smooth, including inner surface of chelal hand, except for rather prominent bases of some setae. Trichobothria on chela as shown in Fig. 10; t short and lanceolate as in I. crassimanum. Palpal femur 2.6-2. 8, tibia 1.85-2.05, chela (without pedicel) 2.35-2.5 times as long as broad; hand (without pedicel) 1.25-1.5 times as long as deep; movable finger 0.83-0.95 as long as hand. Fixed finger with 35-40 and movable finger with 43-49 cusped, marginal teeth. Legs rather stout; leg I with basifemur 1.25-1.3 times as long as telofemur; leg IV with entire femur 2.35-2.55 times as long as deep. Dorsal surface of leg IV depressed at junction between basifemur and telofemur. Subterminal tarsal setae dentate. Tibia, metatarsus and telotarsus each with a prominent tactile seta proximad of middle. Measurements (mm).— Figures given first for the paratype, followed in parentheses by ranges for the other specimens from Guadeloupe and Dominica. Body length 1.58(1.47-2.18). Carapace length 0.55(0.555-0.65). Chelicera 0.32(0.33-0.41) long. Palpal femur 0.45(0.465-0.60) by 0.17(0.185-0.22); tibia 0.415(0.45-0.555) by 0.21(0.235- 0.27); chela (without pedicel) 0.72(0.79-0.985) by 0.295(0.33-0.42); hand (without pedicel) 0.38(0.435-0.585) by 0.27(0.325-0.415); pedicel about 0.07 long; movable finger 0.355(0.39-0.495) long. Leg I: basifemur 0.245(0.235-0.28) long; telofemur 0.18(0.19-0.215) long. Leg IV: entire femur 0.53(0.51-0.59) by 0.23(0.215-0.25); tibia 0.385(0.385-0.445) by 0.095-0.1 1). Remarks.— Though the paratype from St. Vincent is a little smaller than the specimens from Dominica and Guadeloupe, there is a good agreement in other characters and they are all considered conspecific. Ideobisium chapmani, new species Fig. 1 1 Material.— Holotype male (WM 3449.01001) from “guano patches below bat roosts, terminal upstream chamber” of Camburales Cave, 10 km E Curimagua, Serrania de San Luis, Falcon, Venezuela, 20 May 1973; paratype female in “leaf litter on summit of the ridge separating Acarite and Camburales valleys,” 7 km E Curimagua, Serrania de San MUCHMORE-THE GENERA IDEOBISIUM AND IDEOBLOTHR US 199 Luis, Falcon, Venezuela, 20 March 1973; both collections by Philip Chapman [FSCA] . Diagnosis.— Generally similar to I. crassimanum , but appendages a little slimmer (chela of male 2.45 times as long as broad) and trichobothrium eb situated far proximad and ventrad on side of chelal hand. Description.— Male and female generally similar but female larger and darker. Carapace smooth; epistome broad, low rounded; four eyes present, smaller and less convex than in I. crassimanum ; chaetotaxy of holotype 4-6-4-6-7 = 27, of paratype 4-6-4-6-8 = 28. Coxal area typical. Abdominal tergites and posterior sternites entire, anterior sternites weakly divided; pleural membranes heavily granulate anteriorly, becoming longitudinally granulo-striate posteriorly. Tergal chaetotaxy of holotype 6:6:7:8:9:10:9:10:8:7:T1T1T1T:2; sternal chaetotaxy of male 9: [3-3] : (3)4/8(2) : (3)8(3) : 10:11:11:11: 10: 10:T1T:2. Chelicera 0.6 as long as carapace; hand with five setae; flagellum of six or seven setae, the proximal one shorter and simple; fixed finger with about 15 small, and movable finger with about 10 larger, teeth; galea in both sexes long, slender, curved. Palp stout; femur 2. 4-2. 6, tibia 1. 9-2.0, and chela 2.25-2.45 times as long as broad; hand 1.25-1.35 times as long as deep; movable finger 0.85-0.88 as long as hand. All surfaces smooth. Fixed chelal finger with 29-37 and movable finger with 37-45 cusped, marginal teeth. Trichobothria on chela as shown in Fig. 1 1 ; t shorter than the others and lanceolate at the tip; sb somewhat removed from t and st , nevertheless nearer to st than to b\ the three trichobothria on side of hand widely spaced, with isb at base of fixed finger and eb far proximad and ventral. 12 w \ 13 Fig. 10 .-Ideobisium balzanii With: lateral view of right chela. Fig. 11 .-Ideobisium chapmani, new species holotype male: lateral view of left chela. Fig. 12. -Ideobisium peckorum new species, holotype male: lateral view of left chela. Fig. 13 -Ideobisium ecuadorense, new species, holotype male: lateral view of left chela. 200 THE JOURNAL OF ARACHNOLOGY Legs rather short and stout. Leg I with basifemur 1.3-1.35 times as long as telofemur. Leg IV with entire femur 2.6-2.7 and tibia 4.2-4.4 times as long as deep. Tibia, metatarsus and telotarsus of leg IV each with a prominent tactile seta proximad of the middle. Measurements (mm).-Figures given first for the holotype male, followed in paren- theses by those for the female. Body length 1. 7(2.0). Carapace length 0.605(0.725). Chelicera 0.36(0.43) long. Palpal trachanter 0.295(0.37) by 0.18(0.215); femur 0.52(0.635) by 0.215(0.245); tibia 0.495(0.59) by 0.26 (0.295); chela (without pedicel) 0.87(1.06) by 0.355(0.47); hand (without pedicel) 0.49(0.605) by 0.36(0.48); pedicel 0.08(0.09) long; movable finger 0.415(0.53) long. Leg I: basifemur 0.28(0.32) long; telofemur 0.21(0.245) long. Leg IV: entire femur 0.55(0.635) by 0.215(0.235); tibia 0.435(0.525) by 0.105(0.12). Etymology.— The species is named for Philip Chapman, who collected the specimens. Remarks.— The weakly developed eyes of this species might be taken as evidence that it is partially adapted for life in the dark cave. However, other parts of the animals are normal. Further, there are at hand several specimens taken from leaf litter in Valle, Colombia, which have similarly reduced eyes and a similar trichobothrial pattern. These may be conspecific with I. c hap man i, but because they are significantly smaller they have not been considered paratypes. The possibility must be recognized that this species is actually/. (Ideoroncus) gracilis Balzan, which was characterized as having only two eyes (see discussion below). Ideobisium peckorum, new species Fig. 12 Material.— Holotype male (WM2893. 02001) and five paratypes (2d, 39) separated from forest litter 7 km N Leticia, Amazonas, Colombia, 20-25 February 1972 (Stewart and Jarmila Peck), [FSCA] . Diagnosis.— Much like I. chapmani but smaller, with well-developed eyes, with slightly more slender appendages (chela of male 2.75 times as long as broad), and with trichobo- thrium isb distinctly proximad of ib. Description.— Sexes similar though females larger then males. Carapace smooth; epistome very low, broad; four large corneate eyes; chaetotaxy 4-6-4-6-6- = 26. Coxal area typical. Abdominal tergites and posterior sternites entire, anterior sternites weakly divided; pleural membranes strongly granulate anteriorly, becoming longitudinally lined pos- teriorly. Tergal chaetotaxy of holotype male 6:6:6:7:7:8:8:8:7:7:T1T1T1T:2; sternal chaetotaxy of same 8: [3-3] :(3)4/8(3):(3)9(3): 1 2: 11:10: 10:8:8:T1T:2. Anterior genital operculum of female usually with a group of eight setae and posterior operculum with a row of eight setae; one female apparently abnormal in having three small groups of about 10 tiny setae on the anterior operculum and a double row of 11 setae on the posterior operculum. Chelicera nearly 0.6 as long as carapace; hand with five setae; flagellum of seven setae, the proximal one shorter and simple; fixed finger with about 15 small, and movable finger with about 8 larger, teeth; galea in both sexes long, slender, gently curved. Palp stout; femur 2.55-3.0, tibia 1.95-2.05, and chela 2.4-2.75 times as long as broad; hand 1.4-1.55 times as long as deep; movable finger 0.81-0.87 as long as hand. All surfaces smooth. Fixed chelal finger with 32-37 and movable finger with 43-47 cusped, marginal teeth. Trichobothria on chela as shown in Fig. 12; Ms shorter than the others MUCHMORE-THE GENERA IDEOBISIUM AND IDEOBLOTHR US 201 and lanceolate at the tip; sb somewhat removed from t and st , nevertheless nearer to st than to b; the three trichobothria on the side of the hand rather widely spaced, with isb distinctly proximad of ib and eb below and behind the middle of the hand. Legs rather short and stout. Leg I with basifemur 1.25-1.3 times as long as telofemur. Leg IV with entire femur 2.25-2.75 and tibia 3.9-4.05 times as long as deep. Tibia, metatarsus and telotarsus of leg IV each with a prominent tactile seta proximal to the middle. Measurements (mm).— Figures given first for the holotype male, followed in paren- theses by ranges for the five paratypes. Body length 1.72(1.74-2.21). Carapace length 0.605(0.59-0.66). Palpal trochanter 0.28(0.27-0.315) by 0.17(0.165-0.185); femur 0.51(0.48-0.525) by 0.17(0.165-0.205); tibia 0.445(0.43-0.48) by 0.22(0.21-0.245); chela (without pedicel) 0.815(0.80-0.92) by 0.325(0.295-0.38); hand (without pedicel) 0.48(0.45-0.54) by 0.32(0.29-0.385); pedicel about 0.065 long; movable finger 0.39 (0.385-0.445) long. Leg I: basifemur 0.245(0.235-0.27) long; telofemur 0.20(0.19-0.205) long. Leg IV: entire femur 0.529(0.495-0.54) by 0.195(0.18-0.215); tibia 0.385(0.37-0.42) by 0.095-0.105). Etymology.— The species is named for Stewart and Jarmila Peck who collected these as well as many other tropical pseudoscorpions. Ideobisium ecuadorense, new species Fig. 13 Material.— Holotype male (WM 4700.01001) and paratype female from forest litter near Los Tayos Caves, Cordillera el Condor, Ecuador (3°0l'S, 78°15'W), 17 July and 1 August 1976 (N. P. Ashmole), [FSCA] . Diagnosis.— Slightly larger than I. crassimanum, with 26 setae on carapace, and with trichobothria eb and esb near middle of lateral side of chelal hand, both far removed from isb. Description.— Female like male but a little larger. Carapace and palps light reddish brown, other parts lighter. Carapace smooth; epistome very low, rounded; four large, corneate eyes; chaetotaxy 4-6-4-6-6 = 26. Coxal area typical; palpal coxa with two long setae on apex. Abdomen elongate; tergites and sternites entire; surfaces smooth; pleural membranes heavily granulate anteriorly, becoming longitudinally striate on posterior segments. Tergal chaetotaxy of male 6:6:6:8:8:8:8:8:8:7:T1T1T1T:2; sternal chaetotaxy 8: [3-3] :(3)4/ 1 0(3):(2)1 0(3): 10: 1 1 : 10: 10: 9: 9: TIT: 2. Middle and posterior tergites and sternites of female with one or two more setae than in male; anterior operculum with eight setae, arranged as in I. crassimanum. Chelicera about 0.6 as long as carapace; hand with five setae; flagellum of six or seven setae, the proximal one shorter and simple; fixed finger with 14 or 15 small, and movable finger with 11 or 12 larger, teeth; galea simple, longer and more curved in female than in male. Palp stout; femur 2.65-2.9, tibia 1.95-2.05, and chela 2.25-2.5 times as long as deep; hand 1.35-1.4 times as long as deep; movable finger 0.85-0.86 as long as hand. All surfaces smooth. Fixed chelal finger with 30-32 and movable finger with 42-43 cusped marginal teeth. Trichobothria on chela as shown in Fig. 13; t is shorter than the others and lanceolate at the tip; st a little nearer to t than to sb; isb at the same level as ib and well separated from eb and esb which lie near middle of side of hand. 202 THE JOURNAL OF ARACHNOLOGY Legs rather short and stout. Leg I with basifemur 1.3 times as long as telofemur. Leg IV with entire femur 2. 5-2. 7 and tibia 4.0-4. 2 times as long as deep. Tibia, metatarsus and telotarsus each with a prominent tactile seta proximad to the middle. Measurements (mm).-Figures given first for the holotype male, followed in paren- theses by those for the female. Body length 1.84(2.0). Carapace length 0.615(0.62). Palpal trochanter 0.295(0.31) by 0.19(0.20); femur 0.555(0.57) by 0.19(0.215); tibia 0.48(0.51) by 0.245(0.25); chela (without pedicel) 0.85(0.92) by 0.34(0.41); hand (without pedicel) 0.48(0.54) by 0.34(0.40); pedicel about 0.07 long; movable finger 0.415(0.46) long. Leg I: basifemur 0.265(0.29) long; telofemur 0.21(0.22) long. Leg IV: entire femur 0.54(0.58) by 0.20(0.23); tibia 0.42(0.465) by 0.105(0.1 1). Etymology.— The species is named for Ecuador, the country in which it has been found. Ideobisium puertoricense, new species Figs. 14, 15 Material.— Holotype male (WM 2509.02003) and 20 paratypes (13d, 79) from rain- forest litter in the Luquillo Mountains (elev. 424m) in NE Puerto Rico, 28 March 1967, E. W. McMahon, [FSCA] ; one male and two female paratypes from basal “leaf sheaths” of dead cycads on El Yunque beside route 915 in NE Puerto Rico, 18 September 1977, A. R. Gillogly and H. J. Harlan, [FSCA] ; many paratypes (3d, 39 mounted) from litter in Aguas Buenas Forest near Aguas Buenas Cave (elev. 25m) in east central Puetro Rico, 7-17 May 1973, S. B. Peck, [FSCA]. Five specimens (4d, 19) from leaf litter, near Maricao, in western Puerto Rico, 5 January 1977, J. A. Mari Mutt, [FSCA] considered conspecific in spite of some small differences. Diagnosis.— Similar to I. crassimanum but with palpal chela less stout, 1/w of chela 2.2 or greater and movable finger 0.85 or more as long as hand; carapace usually with 26 setae, rather than 24. Description.— Males and females similar, but females larger. Palps and carapace well sclerotized and reddish brown, other parts lighter. Carapace smooth; epistome broad, low, rounded; four eyes; chaetotaxy of holotype 4-6-4-6-6 = 26, others similar but often with 4, 5, or 7 instead of 6 in the fourth row. Coxal area typical. Abdominal tergites and posterior sternites entire, anterior sternites weakly divided; pleural membranes heavily granulate anteriorly, becoming longitudinally granulo-striate posteriorly. Tergal chaetotaxy of holotype male 6:6:6:9:9:8:9:9:7:7:T1T1T1T:2, sternal chaetotaxy of same 9: [3-3] :(3)4/8(3):(3)8(3):l 1:10:10:10:8:9:T1T:2; others similar but variable; anterior genital operculum of female with seven to nine setae. Chelicera about 0.6 as long as carapace; hand with five setae, es short; flagellum of six or seven dentate setae, the proximal one shorter than the others; each finger with six to eight teeth; galea slender, curved, longer in female than in male; serrula exterior of about 30 blades. Palp stout (Fig. 14); femur 2. 5-3.0, tibia 1.85-2.15, and chela 2. 2-2.1 times as long as broad; hand 1.25-1.5 times as long as deep; movable finger 0.85-0.95 as long as hand. All surfaces smooth. Fixed chelal finger with 29-41 and movable finger with 43-58 cusped, marginal teeth. Trichobothria on chela as shown in Fig. 15; t shorter than the others and broadly lanceolate on outer third; esb nearer to eb than to isb on side of hand. MUCHMORE-THE GENERA IDEOBISIUM AND IDEOBLOTHRUS 203 Legs rather short and stout. Leg I with basifemur 1.2-1. 3 times as long as telofemur. Leg IV with entire femur 2.4-2. 8 and tibia 3.65-4.25 times as long as deep. Tibia, metatarsus and telotarsus each with a tactile seta proximad of the middle. Measurements (mm).-Figures given first for the holotype, followed in parentheses by ranges for the mounted paratypes. Body length 1.8(1. 7-2.4). Carapace length 0.63(0.57-0.68). Chelicera 0.35(0.31-0.43) long. Palpal femur 0.55(0.47-0.65) by 0.21(0.18-0.22); tibia 0.495(0.42-0.57) by 0.255(0.275-0.29); chela (without pedicel) 0.85(0.72-1.035) by 0.335(0.30-0.46); hand (without pedicel) 0.465(0.41 - 0.605) by 0.325(0.31-0.45); pedicel 0.05-0.075 long; movable finger 0.42(0.36-0.52) long. Leg I: basifemur 0.265(0.23-0.32) long; telofemur 0.215(0.18-0.25) long. Leg IV: entire femur 0.56(0.50-0.66) by 0.22(0.18-0.235); tibia 0.41(0.36-0.51) by 0.105(0.095-0.12). Etymology .—The species is named for Puerto Rico, where it is found. Remarks.— Three females from the Dominican Republic appear to belong to this species. They were collected by W. L. Brown at La Cienaga, La Vega, at an altitude of 1100m. The I. crassimanum recorded by Beier (1976) from 1250m in the Cordillera Central, Dominican Republic, may belong here. Ideobisium puertoricense cavicolum, new subspecies Material.— Holotype male (WM 3931.01006) and 12 paratypes from bat guano in Aguas Buenas Cave, Aguas Buenas, Puerto Rico, 3 May 1974, S. B. Peck [FSCA] . Figs. 14, 15 .-Ideobisium puertoricense, new species, holotype male: 14, dorsal view of right palp; 15, lateral view of left chela. Fig. 16 .-Ideobisium yunquense, new species, holotype male: lateral view of left chela. 204 THE JOURNAL OF ARACHNOLOGY Diagnosis.— When compared to specimens of /. p. puertoricense found just outside Aguas Buenas Cave, this population differs in several important features which suggest incipient adaptation to the subterranean environment — the color, especially of cara- pace, tergites and legs, is lighter; the size is larger; and the palpal segments are less stout. Description.— Palps reddish brown, but carapace, abdomen and legs much lighter. Carapace like the nominate subspecies, with four eyes; chaetotaxy 4-6A-6-6 = 26. Palpal femur 2.85-3.15, tibia 2.0-2.15, and chela 2.25-2.65 times as long as broad; hand 1.3-1. 5 times as long as deep; movable finger 0.81-0.90 as long as hand. Leg IV with entire femur 2. 6-2. 9 and tibia 4.25-4.8 times as long as deep. Measurements (mm).— Figures given first for holotype, followed in parentheses by ranges for the 11 adult paratypes. Body length 2.1(2.0-2.75). Carapace length 0.665(0.64-0.755). Chelicera 0.385(0.36-0.44) long. Palpal femur 0.63(0.59-0.70) by 0.20(0.20-0.24); tibia 0.58(0.53-0.63) by 0.27(0.27-0.32); chela (without pedicel) 0.94(0.90-1.11) by 0.37(0.36-0.48); hand (without pedicel) 0.52(0.50-0.635); pedicel 0.07-0.08 long; movable finger 0.47(0.44-0.52) long. Leg I: basifemur 0.27(0.265-0.33) long; telofemur 0.22(0.22-0.265). Leg IV: entire femur 0.62(0.60-0.70) by 0.23(0.22- 0.25); tibia 0.47(0.46-0.56) by 0.1 1(0.10-0.13). Etymology.— The subspecies is named cavicolum for its cavernicolous habitat. Ideobisium yunquense, new species Fig. 16 Material.— Holotype male (WM 2250.01001) and paratype female from Mt. Britton, El Yunque (elev. 730m) in NE Puerto Rico, 6 September 1964, [FSCA] . Paratype male and female from El Yunque, Puerto Rico, April 1969, (T. Hlavac), [FSCA]. Paratype male from El Yunque Biological Station (elev. 825m), Puerto Rico, 25 Jan- uary 1964, [MCZ] . Diagnosis.— Similar to I. puertoricense but larger (palpal femur longer than 0.70 mm) and with slightly more slender appendages (palpal femur with 1/w 3.0 or greater). Description.— Male and female similar but female larger. Palps, carapace and tergites dusky brown, other parts lighter. Carapace smooth; epistome low, broad, rounded; four eyes; chaetotaxy of holotype and one paratype 4-6-4-6-6 =26, two paratypes with 4-6-4-5-6 = 25. Coxal area typical. Abdominal tergites and posterior sternites entire, anterior sternites divided; pleural membranes heavily granulate anteriorly, becoming longitudinally granulo-striate poste- riorly. Tergal chaetotaxy of holotype 6:6:6:8:9:9:9:9:7:7:T1T1T1T:2; sternal chaeto- taxy of same 7: [3-3] :(3)4/ll(3):(3)8(3):12: 11:1 1:10: 10: 10:T1T:2; others similar but varied; anterior operculum of female with seven or eight setae. Chelicera about 0.6 as long as carapace; hand with five setae, b and es shorter than others; each finger with 8-12 teeth; galea slender, curved, longer in female than in male; serrula exterior with about 32 blades. Palp rather stout; femur 3.0-3.05, tibia 2.05-2.15, and chela 2.4-2. 7 times as long as broad; hand 1.35-1.5 times as long as deep; movable finger 0. 9-1.0 as long as hand. All surfaces smooth. Fixed chelal finger with 38-40 and movable finger with 50-54 cusped teeth. Trichobothria on chela as shown in Fig. 16; t broadly lanceolate in the outer 2/5; esb nearer to eb than to isb on side of hand. MUCHMORE-THE GENERA IDEOBISIUM AND IDEOBLOTHRUS 205 Legs rather short and stout. Leg I with basifemur 1.25-1.35 times as long as telo- femur. Leg IV with entire femur 2.65-2.85 and tibia 4. 2-4.4 times as long as deep. Tactile seta proximad of middle on tibia and each tarsal segment. Measurements (mm).— Figures given first for the holotype, followed in parentheses by ranges for the three paratypes. Body length 2.15(2.2-2.7). Carapace length 0.76(0.755-0.83). Chehcera 0.445(0.41-0.495) long. Palpal femur 0.70(0.70-0.755) by 0.23(0.23-0.25); tibia 0.635(0.63-0.665) by 0.30(0.295-0.325); chela (without pedicel) 1.065(1.095-1.21) by 0.42(0.41-0.52); hand (without pedicel) 0.59(0.585-0.71) by 0.40(0.39-0.53); pedicel about 0.07 long; movable finger 0.555(0.58-0.635) long. Leg I: basifemur 0.33 (0.34-0.36) long; telofemur 0.265(0.26-0.265) long. Leg IV: entire femur 0.73(0.70-0.77) by 0.275(0.25-0.27); tibia 0.55(0.555-0.585) by 0.13(0.125-0.14). Etymology .-The species is named for the mountain, El Yunque, on which it is found. Ideobisium peregrinum Chamberlin Ideobisium peregrinum Chamberlin 1930:37, 1931:Figs. 11R, 35F, G, H, 40S, 43J, K, 50C. Material examined.— Holotype male (JC 94.02001), from under long in beech forest at Days Bay, Wellington Harbor, New Zealand [JCC] ; allotype female, in leafmould near Wellington, New Zealand [JCC] ; paratype female from Kingston, Lake Watipu (= Wakati- pu), New Zealand, [BM(NH)] . Diagnosis.— A good species of Ideobisium , much like the American forms, but with eight setae at posterior margin of carapace and eight or more setae on tergite 2. Description.— The original description by Chamberlin is very brief and without illus- tration; however, several figures of the species were provided later (Chamberlin 1930: listed above). In order to compare this species with others in the genus, the following supplemental data are given. Male and female much alike. Palps, carapace and tergites well sclerotized, brown. Carapace smooth; epistome low, broad, rounded; four eyes, with flattened corneas; chaetotaxy of holotype 4-44-5-8 = 25, of allotype 4-4-44-7 = 23, and of paratype 4-44-3-8 = 23. Coxal area typical; palpal coxa with two long setae at the apex. Abdominal tergites and posterior sternites entire, anterior sternites weakly divided; pleural membranes granulate anteriorly, becoming granulo-striate posteriorly. Tergal chaetotaxy of holotype male 6:8:9:9:9:10: 10: 10: 8:7:T1T1T1T:2; sternal chaetotaxy of same 8: [3-3] : (4)6(3): (2)6(2) :1 1:10:12:12:1 1:11 :T1T:2; anterior operculum of female with six setae in a transverse row. Internal genitalia of male shown by Chamberlin (1931 : Fig. 50C). Chelicera 0.6 as long as carapace; hand with five setae, es short; flagellum of six or seven dentate setae, the proximal one shorter than the others; each finger with 10-15 teeth; galea slender, gently curved, longer in female than in male. Palp stout; femur 2.7-2.75, tibia 1.85-2.0, and chela 2. 5-2.6 times as long as broad; hand 1.35-1.5 times as long as deep; movable finger 0.9 as long as hand. All surfaces smooth. Fixed chelal finger with 31-36 and movable finger with 38-42 contiguous marginal teeth, of which only the distal 10-12 are cusped. Trichobothria on chela as shown by Chamberlin (1931 ; Fig. 35G); isb, esb and eb closer together and more distal on the hand than in the American species of the genus; t shorter than the others and lanceolate toward the tip. 206 THE JOURNAL OF ARACHNOLOGY Legs rather short and stout (see Chamberlin 1931 : Figs. 43J, K). Leg I with basifemur 1 .4 times as long as telofemur. Leg IV with entire femur 2.55-2.7 and tibia 3.85-4.0 times as long as deep. Tactile seta proximad of middle on tibia and each tarsal segment. Measurements (mm).— Figures given in order for holotype, allotype, and paratype. Body length 2.18, 2.53, 2.21. Carapace length 0.70, 0.68, 0.66. Chelicera 0.41, 0.42, 0.39 by 0.215, 0.215, 0.20. Palpal femur 0.65, 0.63, 0.57 by 0.235, 0.23, 0.21; tibia 0.605, 0.57, 0.57 by 0.30, 0.295, 0.26; chela (without pedicel) 1.01, 1.035, 0.91 by 0.40, 0.40, ?; hand (without pedicel) 0.585, 0.59, 0.495 by 0.39, 0.40, 0.37; pedicel about 0.08 long; movable finger 0.53, 0.525, 0.46 long. Leg 1: basifemur 0.30, 0.29, 0.26 long; telofemur 0.21, 0.21, 0.185 long. Leg IV: entire femur 0.595, 0.59,0.51 by 0.23, 0.23, 0.19; tibia 0.46, 0.46, 0.40 by 0.12, 0.12, 0.10. Remarks.— It should be noted that the paratype female, from South Island, is smaller than the other two specimens, but is a fully mature female, not immature as stated by Chamberlin (1930:37). The species has also been recorded from Oamaru, on seashore (Beier 1948) and from Nelson, Buller Gorge, No thofagus litter (Beier 1967). Ideobisium antipodum (Simon) Obisium antipodum Simon 1880:174. Ideobisium antipodum : Beier 1932:160, 1940:168, 169, 1968:762, Fig. 4. The original description, based on the type from Noumea, New Caledonia, was supplemented by Beier (1968) on the basis of a male specimen from the Grotte de Ninrin-Reu, near Poya, New Caledonia. It has also been reported from the Ellice Islands (Beier 1940). Ideobisium (?) gracile Balzan Ideobisium (Ideoroncus) gracilis Balzan 1891:540, Fig. 31, 31a; Beier 1932:158. This species is reported only from Venezuela. The type(s), which probably should be in the MNHN, Paris, along with those of Ideobisium crassimanum Balzan and Ideoblo- thrus similis Balzan, cannot be located (per Dr. J. Heurtault). No other specimens are known which conform to the description given by Balzan. Balzan placed I. gracilis in the subgenus Ideoroncus because it purportedly possessed two eyes. However, if his description and figure 31 are accurate, the species does not belong in Ideoroncus or any of the Ideoroncidae as presently understood; the shape of the palps and the epistome on the carapace are more characteristic of Ideobisium or Ideoblothrus. It is possible that this is an Ideobisium like I. chapmani, in which the eyes are reduced and which appeared, on superficial examination, to have only two (anterior) eyes. Final determination in this matter must await location of the type or collection of topotypes. Other species, originally placed in Ideobisium by their authors, have already been disposed of as follows: /. minutum Tullgren 1905, assigned to Hya Chamberlin (Beier 1932:167). I. quadrispinosum Tullgren 1907, made the type species of Gymnobisium Beier (1931:304). MUCHMORE-THE GENERA IDEOBISIUM AND IDEOBLOTHR US 207 I tibiale Banks 1909, tentatively assigned to Syarinus (Chamberlin 1930:40; Beier 1932:164) or Microcreagris Balzan (Hoff 1956:8). Having examined the holo- type female of this species [MCZ] , I support Hoffs assignment. I. magnum Banks 1909, assigned to Microcreagris (Chamberlin 1930:28). I. tacomense Ellingsen 1909a and /. pyrenaicum Ellingsen 1909b, assigned to Microcreagris (Beier 1932:154, 156). I. hispanicum Ellingsen 1910, assigned to Microcreagris (Beier 1932:155), later transferred to Microcreagrina Beier (Beier 1970:45). I. racovitzai Ellingsen 1912b, made the type species of Troglobisium Beier (1939:189). I. orientale Redikorzev 1922, made the type species of Halobisium Chamberlin (1930:35). I. cavimanum Beier 1930, made the type species of Mirobisium Beier (1931:304). I. formosum Mello-Leitao 1937, assigned to Geogarypus Chamberlin (Mahnert 1979:750,761). Genus? formosanum (Ellingsen) Ideobisium formosanum Ellingsen 19 12a: 125; Beier 1932:160. This species is known only from the type specimens from Koroton, Formosa. Though no illustrations are given, Ellingsen is very explicit about the form of the cheliceral galea of this species; it is said to be “shaped like a fan,” with several branches. Because all known species of Ideobisium (and Ideoblothrus ) as here defined have simple galeae, this form must belong to some other genus, possibly to Microcreagris or Syarinus. Obisium trifidum Stecker 1875 was tentatively assigned to Ideobisium by Beier (1932:160); this assignment has been reported without comment by Murthy and Ananthakrishnan (1977:2). No further data on the species have become available, but if Stecker’s figures are correct it cannot be retained in Ideobisium as here defined. The cheliceral galea is quite different from that found in Ideobisium species. Ideoblothrus Balzan Ideoblothrus Balzan 1891:541 (subgenus of Ideobisium Balzan). Type species Ideoblothrus similis Balzan 1891, by original designation. Pachychitra Chamberlin 1938:111. Type species Pachychitra maya Chamberlin 1938, by original designation. Hoff 1945:1; Hummelinck 1948:63; Hoff 1964:7; Muchmore 1972:262, 1979:195. NEW SYNONYMY. Diagnosis (revised).— A genus of the superfamily Neobisioidea Chamberlin (1930:7). Species small and robust. Carapace about square; anterior margin with a small, rounded, triangular epistome; surface smooth; no eyes present; 22-26 vestitural setae, with four at anterior and four to six at posterior margin. Apex of palpal coxa acute, with two long setae. Tergites and sternites entire, except that sternites 3-5 may be weakly divided; surfaces smooth; tergite 1 usually with six or seven setae, following tergites with eight or nine; pleural membranes finely, longitudinally striate. Internal genital setae of male arranged as two triangular groups of three. Cheliceral fingers toothed; hand with five acuminate setae, es shortest; flagellum of five or six finely denticulate setae, and in some 208 THE JOURNAL OF ARACHNOLOGY species a short, simple, proximal one; galea long, simple. Palp robust, none of the segments more than 3.0 times as long as broad; surfaces smooth except for fine granules on flexor sides of femur, tibia, and chelal hand at base of fingers; movable chelal finger usually as long as or longer than chelal hand; venom apparatus developed only in fixed finger, with short duct, extending less than half the distance to trichobothrium et\ chelal teeth contiguous, at least the distal ones cusped. Trichobothrium t on movable chelal finger shorter than all others and lanceolate toward tip. t, st and sb closely grouped in an obliquely oriented series, with t distad of middle of finger; est, it and ist all near middle of fixed finger, distinctly proximad of et; isb, esb and eb in an oblique series on side of hand just proximad of base of fingers. Leg I with basifemur slightly longer (usually 1 .1 or less) than telofemur. Leg IV with dorsal margin of femur smooth across the suture between its parts, the suture itself slightly oblique to long axis of femur; tibia with a tactile seta at or distad of middle, tarsal segments each with a tactile seta proximad of middle. Subterminal tarsal setae finely denticulate in distal half; arolia as long as or longer than claws. Distribution— Northern South America, Central America and Mexico, Greater Antilles and Florida, Central and South Africa, Seychelles Islands, New Guinea, and Solomon and Caroline islands. It is undoubtedly pantropical. Remarks.— Examination of the holotype of Ideoblothrus similis (see below) makes it perfectly clear that Ideoblothrus Balzan and Pachychitra Chamberlin are identical. Ideoblothrus is very similar to Ideobisium Balzan but can easily be distinguished from it by the absence of eyes, and the entirely smooth, longitudinally striate abdominal pleural membranes. In addition, the two genera differ in the detailed location of trichobothria on the chelal fingers, the presence or absence of granules on the palpal segments, the nature of the articulation between parts of the femur of leg IV, the length of the pedal arolia with respect to the claws, and the placement of the internal genital setae of the male. Ideoblothrus also appears similar in general form to Alocobisium Beier , Micro creagrella Beier, and Micro creagrina Beier, but may be separated easily from those genera, all of which have at least one trichobothrium situated well back on the dorsum of the chelal hand. Ideoblothrus similis Balzan Fig. 17 Ideobisium (Ideoblothrus) similis Balzan 1891:541. Ideobisium simile: Beier 1932:159. Material.— Holotype male from Petare, Venezuela (Col. Mus. 13.839), [MNHN] . The specimen is mounted on two microscope slides, numbered 440 and 441; the left chela is missing and the right palpal segments are badly crushed. No other specimens are known. Description of holotype.— All parts light brown or tan. Carapace smooth; epistome broken, but apparently low, rounded; no eyes present; chaetotaxy 4-4-4-4-6 = 22. Coxal area generally typical of the Neobisioidea; apex of palpal coxa acute, with two long setae. Abdomen elongate; tergites and posterior sternites entire, anterior sternites weakly divided; surface smooth; pleural membranes longitudinally striate, the striae irregular anteriorly. Tergal chaetotaxy 7:7:8:9:9:9:9:9:9:7:T1T1T1T:2; sternal chaetotaxy 8: [3-3] :(2)4/6(2):(2)8(2): 10: 11:1 1 :9:9:9:T1T;2; genital opercula as shown in Fig. 17; internal genitalia partly obscured, but as far as can be seen are much like those figured for Pachychitra floridensis Muchmore (1979:196). MUCHMORE-THE GENERA IDEOBISIUM AND IDEOBLOTHR US 209 Chelicera 0.56 as long as carapace; hand with five acuminate setae, es very short; flagellum apparently of five subequal, finely denticulate setae (position is such that is it not possible to determine whether a short proximal seta is present); fixed finger with about 14 small teeth, and movable finger with seven medium and about 10 tiny basal teeth; galea slender, straight; serrula exterior with about 24 blades. Palp rather stout (see Balzan 1891: Fig. 32); segments more or less as described by Balzan, except that movable finger of chela is not “manu breviores”, but is actually 1.1 times as long as hand (without pedicel). Fixed chelal finger with about 35 and movable finger with about 40 contiguous marginal teeth. The chela is badly crushed, but it can be seen that the trichobothria occupy positions as in Pachychitra may a (see Chamberlin 1938:1 10 and Fig. IB); on movable finger t is distal to middle of finger; on fixed finger est, it and ist are near middle of finger distinctly proximad of et; and isb, esb and eb form an oblique linear series on side of hand just proximad of base of fingers; as t is missing from its socket on the one chela present, its form cannot be determined. Legs rather short and stout. Leg I with basifemur only slightly longer than telofemur. Leg IV with entire femur 2.5 and tibia 4.05 times as long as deep. The femora are not in suitable positions to observe the dorsal outline or the interfemoral suture. Subterminal setae finely denticulate; arolia slightly longer than claws. Tibia of leg IV with a tactile seta just distad of middle; each tarsal segment with a tactile seta proximad of middle. Measurements (mm).— Body length 1.62. Carapace length 0.435. Chelicera 0.245 by 0.125. Palpal femur 0.385 by ?; tibia 0.37 by ?; chela (without pedicel) 0.63 by ?; hand (without pedicel) 0.31 by ?; pedicel 0.06 long; movable finger 0.34 long. Leg I: basifemur Fig. 17 .-Ideoblothrus similis Balzan, holotype male: genital opercula. Figs. 18-20 -Ideoblothrus kochalkai, new species, holotype male: 18, lateral view of left chela; 19, leg IV: 20, subterminal tarsal seta (much enlarged). 210 THE JOURNAL OF ARACHNOLOGY 0.16 long; telofemur 0.145 long. Leg IV: entire femur 0.31 by 0.125; tibia 0.265 by 0.065; metatarsus 0.095-0.05; telotarsus 0.155 by 0.04. Remarks.— It is unfortunate that trichobothrium t is missing from its socket on the one chela available for this specimen. However, it can reasonably be expected that its form would be like that in Pachychitra species because this species resembles Pachychitra maya and others in all major details. Beier (1974:101) considered Ideobisium costaricense (see below) a synonym of I. similis. This is almost certainly incorrect, given the great geographical separation of the two forms and the large amount of speciation in the genus. Ideoblothrus kochalkai , new species Figs. 18-20 Material.— Holotype male (WM 4837.01001) and paratype female from under rocks at Casa Antonio (elev. 2700 m), Cuchilla Cebolleta, Sierra Nevada de Santa Marta, Mag- dalena, Colombia, 8 May 1975, J. A. Kochalka [FSCA] . Diagnosis.— Generally similar to I. similis, but much larger and with more slender appendages; palpal femur longer than 0.7 mm and with 1/w greater than 2.75. Description.— Male and female similar but female larger. Carapace and palps reddish brown, other parts lighter. Carapace smooth; epistome small, triangular; no eyes;chaeto- taxy 4-4-4-4-5 = 21 . Coxal area typical; palpal coxa acute, with two long setae. Abdominal tergites and posterior sternites entire, anterior sternites weakly divided; pleural membranes longitudinally striate. Tergal chaetotaxy of holotype 6:8:9:9:9:9: 9:9:9:7:T1T2T1T:2; sternal chaetotaxy (male) 15: [3-3] :(3)4/6(3):(2)9(2): 10:7:9:10:10:9: T1T:2; anterior genital operculum of female with transverse row of six setae. Chelicera 0.6 as long as carapace; hand with five setae, es very short; flagellum of six or seven denticulate setae, the proximal one shorter than the others; galea of male slender, straight, not reaching as far as tip of galeal seta, that of female curved, reaching beyond gs; serrula exterior with about 30 blades. Palp stout; femur 2.8-3.05, tibia 2.1, and chela 2.55-2.9 times as long asbroad;hand 1.4-1.45 times as long as deep; movable finger about as long as hand. Surfaces smooth except for a few granules on medial sides of femur, tibia, and chela at base of fingers. Fixed chelal fingers with 47-48 contiguous, cusped, marginal teeth; movable finger with 55-57 contiguous teeth, only the distalmost 10-12 with cusps. Trichobothria on chela as shown in Fig. 18; t shorter than the others and narrowly blade-like in the outer third. Legs less stout than usual for the genus (Fig. 19); leg I with basifemur 1.05-1.1 times as long as telofemur; leg IV with entire femur 3. 0-3. 2 and tibia 4.45-4.75 times as long as deep. Tactile seta on leg IV just distad of middle of tibia and proximad of middle of each tarsal segment. Subterminal tarsal setae strongly dentate on distal half (Fig. 20). Measurements (mm).— Figures given first for the holotype male, followed in paren- theses by those of paratype female. Body length 2.3(2.85). Carapace length 0.74(0.88). Chelicera 0.445(0.53) long. Palpal femur 0.72(0.88) by 0.255(0.29); tibia 0.665(0.78) by 0.32(0.385); chela (without pedicel) 1.18 (1.43) by 0.41(0.555); hand (without pedicel) 0.59(0.77) by 0.41(0.54); pedicel 0.10(0.12) long; movable finger 0.66(0.725) long. Leg I: basifemur 0.30(0.34) long; telofemur 0.27(0.32) long. Leg IV: entire femur 0.695(0.785) by 0.23 (0.245); tibia 0.555(0.64) by 0.125(0.135). Etymology.— The species is named for John A. Kochalka, who collected the specimens. MUCHMORE-THE GENERA IDEOBISIUM AND IDEOBLOTHR US 211 Ideoblothrus colombiae, new species Figs. 21, 22 Material.— Holotype male (WM 4838.01002) and four paratypes (3d, 19) sifted from leaf litter between San Pedro and San Javier (elev. 1563 m), Sierra Nevada de Santa Marta, Magdalena, Colombia, 29 March 1975, J. A. Kochalka; paratype male from leaf litter N of San Pedro (elev. 960 m), S. N. de Santa Marta, 19 May 1975, J. A. Kochalka, [FSCA] . Diagnosis.— Generally similar to /. similis but slightly larger (palpal femur 0.41 mm or longer) and with fewer setae on carapace and abdomen. Description.— Male and female similar but female slightly larger. Carapace and palps light brown, other parts much lighter. Carapace smooth; epistome small, rounded; no eyes; chaetotaxy of holotype and three paratypes 4-4-4-4-5 = 21, others with only 4 at posterior margin. Coxal area typical; palpal coxa acute, with two long setae. Abdominal tergites and posterior sternites entire, anterior sternites faintly divided; pleural membranes longitudinally striate, the striae slightly roughened anteriorly. Tergal chaetotaxy of holotype 6:7:9:8:9:8:8:8:9:7:T1T1T1T:2; others similar; sternal chaeto- taxy of holotype male 9: [3-3] :(3)4/5(3):(2)8(2): 10:1 0:9 :9:9:8:T1T:2, other males sim- ilar; anterior genital operculum of female with seven setae in a transverse row. Chelicera about 0.55 as long as carapace; hand with five setae, es very short; flagellum of six denticulate setae, the proximal one only half as long as the others; galea of male slender, straight, not reaching as far as tip of galeal setae, that of female curved, reaching beyond gs; serrula exterior with about 28 blades. Palp stout (Fig. 21); femur 2.3-2.6, tibia 1. 9-2.0, and chela 2.55-2.75 times as long as broad; hand 1.3-1.35 times as long as deep; movable finger 1.03-1.17 times as long as Figs. 21, 22 .-Ideoblothrus colombiae , new species, holotype male: 21, dorsal view of left palp; 22, lateral view of right chela. Figs. 23, 24 -Ideoblothrus seychellesensis (Chamberlin), holotype: 23, lateral view of left chela; 24, subterminal tarsal seta (much enlarged). 212 THE JOURNAL OF ARACHNOLOGY hand. Surfaces smooth except fine granulations medially on femur and tibia and distinct granules on chela at base of fingers. Fixed chelal finger with 31-34 contiguous, cusped marginal teeth; movable finger with 37-40 contiguous teeth, only the distalmost 8-12 cusped. Trichobothria on chela as shown in Fig. 22; t shorter than others and narrowly flattened in outer third. Legs rather stout; leg I with basifemur 1. 0-1.1 times as long as telofemur; leg IV with entire femur 2. 5-2. 7 and tibia 2.55-2.7 times as long as deep. Tactile setae on leg IV just distad of middle of tibia and proximad of middle of each tarsal segment. Subterminal tarsal setae strongly dentate in outer halves; arolia slightly longer than claws. Measurements (mm).— Figures given first for holotype, followed in parentheses by ranges for the paratypes. Body length 1.75(1.7-1.8). Carapace length 0.55(0.52-0.56). Chelicera length 0.30(0.27-0.30). Palpal femur 0.46(0.41-0.48) by 0.18(0.16-0.185); tibia 0.435(0.40-0.445) by 0.22(0.20-0.23); chela (without pedicel) 0.725(0.67-0.80) by 0.27(0.245-0.31); hand (without pedicel) 0.37(0.34-0.40) by 0.27(0.25-0.30); pedicel about 0.07 long; movable finger 0.41(0.38-0.41). Leg I: basifemur 0.18(0.17-0.185) long; telofemur 0.18(0.16-0.17) long. Leg IV: entire femur 0.43(0.41-0.47) by 0.165(0.155-0.18); tibia 0.325(0.32-0.36) by 0.09(0.085-0.095). Etymology.— The species is named for Colombia, where it is found. Remarks.— This species is similar to I. kochalkai but is much smaller — length of palpal femur 0.41-0.48 mm compared to 0.72-0.88 mm. Ideoblothrus seychellesensis (Chamberlin), new combination Figs. 23, 24 Ideobisium seychellesensis Chamberlin 1930:38, Figs. IX, DD, 2CC, Beier 1932:160. Ideobisium seychellesense : Beier 1940:165; Mahnert 1978b:885. The holotype (JC 510.01001) has been examined. As Chamberlin (1930:38) has explained, the body was lost and the specimen presently consists of two palps, two chelicerae, two legs I and two legs IV, mounted on two slides [BM(NH), 1924-X1-3.49 and 3. 49 A] . This, the only known specimen, was taken by the Seychelles Expedition of 1908, probably on Felicite Island, Seychelles Islands (not “on the Felicete” as Chamber- lin, states). For some unknown reason the type slides are labelled Xenobisium seychel- lesensis, in Chamberlin’s hand. Evidently, Chamberlin contemplated erecting a genus Xenobisium to include this species, but he never actually did so. As Chamberlin’s description is rather brief, a more detailed account is given here. Chamberlin reported that the specimen is a female; in the absence of the abdomen, this cannot be verified. Chelicera with five setae on the hand, es rather short; flagellum of six setae, subequal in size and all finely denticulate; fixed finger with 11 small, and movable finger with 9 larger, teeth; galea slender, curved and reaching beyond tip of galeal seta. Palp stout ; femur 2.6, tibia 1 .9, and chela 2.3 times as long as broad; hand 1 .3 times as long as deep; movable finger 0.96 as long as hand. Surfaces smooth except sparse, fine granulation on medial sides of femur, tibia and chela at base of fingers. Fixed chelal finger with 34 and movable finger with 42 contiguous teeth, in each case only the more distal ones cusped. Trichobothria on chela as shown in Fig. 23; t much shorter than the others and thickened, but apparently not flattened (both are present). Legs rather short and stout, as shown by Chamberlin (1930:Figs 1, X and DD). Leg I with basifemur 1.15 times as long as telofemur. Leg IV with entire femur 2.55 and tibia MUCHMORE-THE GENERA IDEOBISIUM AND IDEOBLOTHRUS 213 3.5 times as long as deep; outer contour of femur smooth; tactile setae just distal of middle on tibia, but proximal of middle on each tarsal segment. Pedal arolia slightly shorter than claws; subterminal tarsal setae strongly denticulate near tip, as shown in Fig. 24. Measurements (mm).— Chelicera 0.27 by 0.15. Palpal trochanter 0.25 by 0.19; femur 0.39 by 0.15; tibia 0.37 by 0.195; chela (without pedicel) 0.65 by 0.285; hand (without pedicel) 0.355 by 0.275; pedicel 0.05 long; movable finger 0.34 long. Leg I: basifemur 0.17 long; telofemur 0.15 long. Leg IV: entire femur 0.38 by 0.15; tibia 0.28 by 0.08; metatarsus 0.095 by 0.06; telotarsus 0.155 by 0.05. Remarks.— In some details this description differs from that of Chamberlin. Most of the differences are small and of no significance. However, it should be noted that the palps are not entirely smooth, as suggested by Chamberlin, but actually do have a few fine granules on the femur, tibia and chela. Also, the movable chelal finger is not “clearly a little longer than hand,” as stated by Chamberlin, but is actually a little shorter than the hand. And further, the subterminal tarsal setae are not quite as depicted by Chamberlin (1930: Fig. 2, CC); they are more as shown in Fig. 24, thus not so different from those of other species of Ideoblothrus (cf. Chamberlin 1938: Fig. 1 , F; and Fig. 20 above). In spite of the absence of the body showing the character of the eyes and pleural membranes, it is evident that this species belongs not to Ideobisium but to Ideoblothrus. This is evidenced by the granulations on the palpal segments, the disposition of trichobo- thria on the chela, and the smooth contour of femur IV. It should be noted that Mahnert’s (1978) implication that the pleural membrane of this species is granulate is probably incorrect. Chamberlin did not record the character of the pleural membranes before losing the abdomen, and no other specimens have been examined. However, it is likely that the pleural membranes here are smoothly striate, as in other members of the genus. The following species are assigned to Ideoblothrus because, according to detailed descriptions, they possess combinations of characters clearly representative of that genus, e.g., lack of eyes, longitudinally striate pleural membranes, robust palps, distribution of trichobothria on palpal chela, and acute apex of palpal coxa with two setae. Ideoblothrus costaricensis (Beier), new combination Ideobisium costaricense Beier 1931:302, Fig. 2, 1932:159, Fig. 191. Beier (1974) placed this species in the synonymy of I. similis, but it is almost certainly distinct. Distribution.— Known only from the type localities, Tuis and San Jose, Costa Rica. Ideoblothrus maya (Chamberlin), new combination Pachychitra maya Chamberlin 1938:111, Figs. 1A-F; Hummelinck 1948:71, Figs. 26a-h, 27a-c. I have examined the holotype female of Pachychitra maya (JC 897.01001), [JCC] . It is clearly congeneric with I. similis. Distribution.— Known only from the type locality, “first cave on San Roque Road”, Oxkutzcab, Yucatan, Mexico. 214 THE JOURNAL OF ARACHNOLOGY Ideoblothrus fenestratus (Beier), new combination Ideobisium fenestration Beier 1954:3, Fig. 3. Distribution. -Known only from the type locality, Sivia, south Peru. Ideoblothrus mexicanus (Muchmore), new combination Pachychitra mexicana Muchmore 1972:262, Figs. 1-3. Distribution.— Tamaulipas, Mexico. Ideoblothrus vampirorum, new name Pachychitra similis Muchmore 1972:264, Figs. 4, 5. nec Ideoblothrus similis Balzan 1891:541. With the recognition that Pachychitra is a synonym of Ideoblothrus, the name Pachychitra similis Muchmore becomes a junior homonym of Ideoblothrus similis Balzan and must be replaced. The new name vampirorum refers to the vampire bats in Cueva de los Vampiros, the type locality for the species. Distribution.— Known only from the type locality in Tamaulipas, Mexico. Ideoblothrus grandis (Muchmore), new combination Pachychitra grandis Muchmore 1972:266, Figs. 6, 7. Distribution.— Known only from the type locality, Cueva del Tio Ticho, Chiapas, Mexico. Ideoblothrus insularum (Hoff), new combination Pachychitra insularum Hoff 1945:1, Figs. 1-5: Hummelinck 1948:73, Figs. 29a-b; Hoff 1964:8. Distribution.— Desecheo Is., Puerto Rico, and Jamaica. Ideoblothrus curazavius (Hummelinck), new combination Pachychitra curazavia Hummelinck 1948:63, Figs. 23a-g, 24a-f, 25a-f. Distribution.— Known only from Curasao. Ideoblothrus pygmaeus (Hoff), new combination Pachychitra pygmaea Hoff 1964:9, Figs. 1, 2. Distribution .—Jamaica . MUCHMORE-THE GENERA IDEOBISIUM AND IDEOBL O THR US 215 Ideoblothrus truncatus (Hoff), new combination Pachychitra truncat a Hoff 1964:11, Figs. 3, 4. Distribution. -Known only from the type locality, Maggotty Falls, Parish of St. Elizabeth, Jamaica. Ideoblothrus carinatus (Hoff), new combination Pachychitra carinata Hoff 1964:13, Figs. 5, 6a-b. Distribution.— Known only from the type locality, near Hardwar Gap, Parish of St. Andrew, Jamaica. Ideoblothrus floridensis (Muchmore), new combination Pachychitra floridensis Muchmore 1979:195, Figs. 1-6. Distribution.— Known only from the type locality, Big Pine Key, Monroe County, Florida. Ideoblothrus amazonicus (Mahnert), new combination Ideobisium amazonicum Mahnert 1979:743, Figs. 48-52. Distribution.— Known only from the type locality, Rio Demeni, northern Amazonas, Brasil. Ideoblothrus caecus (Mahnert), new combination Ideobisium caecum Mahnert 1979:745, Figs. 53-58. Distribution.— Known from along the Amazon River between Santarem and Manaus, Brasil. Ideoblothrus brasiliensis (Mahnert), new combination Ideobisium brasiliense Mahnert 1979:747, Figs. 59-64. Distribution.— Known from along the Amazon River between Belem and Manaus, Brasil. Ideoblothrus godfreyi Ellingsen, new combination Ideobisium (Ideoblothrus) godfreyi Ellingsen 1912c: 117; Beier 1947:290; Mahnert 1978a:94. Gymnobisium godfreyi: Beier 1932:162. Distribution.— Known only from the type locality, Frankfort Hill, King Williams Town Div., Cape Province, South Africa. 216 THE JOURNAL OF ARACHNOLOGY Ideoblothrus lepesmei (Vachon), new combination Ideobisium lepesmei Vachon 1941:32, Mahnert 1978a:94. Distribution.— Known only from the type locality, Sassandra, Ivory Coast, West Africa. Ideoblothrus holmi (Beier), new combination Ideobisium holmi Beier 1955:534, Fig. 5; Mahnert 1978a:94. Distribution.— Known from eastern Kenya and western Democratic Republic of the Congo. Ideoblothrus leleupi (Beier), new combination Ideobisium leleupi Beier 1959:23, Fig. 10; Mahnert 1978a:94. Distribution.— Known only from the type locality in western Democratic Republic of the Congo. Ideoblothrus occidentalis (Beier), new combination Ideobisium occidentale Beier 1959:25, Fig. 11; Mahnert 1978a:94. Distribution.-Known only from the type locality along the lower Congo River in western Democratic Republic of the Congo. Ideoblothrus baloghi (Mahnert), new combination Ideobisium baloghi Mahnert 1978a:40, Figs. 48-50. Distribution.-Known only from the type locality near Kindamba, Congo-Brazzaville, West Africa. Ideoblothrus zicsii (Mahnert), new combination Ideobisium zicsii Mahnert 1978a:92, Figs. 51-55. Distribution.-Known only from the type locality near Sibiti, Congo-Brazzaville, West Africa. Ideoblothrus bipectinatus (Daday) Ideobisium bipectinatum Daday 1897:478, Tab, XI, Figs. 7, 14, 15; Beier 1932:160, Fig. 192; 1940:167; Morikawa 1963:4, Figs. 2a-c; Beier 1965:761, 1967:321. Ideoblothrus bipectinatus : With 1906:87. Distribution.— New Guinea and the Bismarck Archipelago. MUCHMORE-THE GENERA IDEOBISIUM AND IDEOBLOTHRUS 217 Ideohlothrus palauensis (Beier), new combination Ideobisium palausense Beier 1957:13, Fig. 4a. Distribution. -Known only from tye type locality, Palau, Caroline Islands. Ideohlothrus pugil pugil (Beier), new combination Ideobisium pugil Beier 1964:593, Fig. 1, 1966:137. Distribution.— Solomon Islands. Ideohlothrus pugil rohustus (Beier), new combination Ideobisium pugil robustum Beier 1966:137, Fig. 3. Distribution.— Known only from the type locality, Nila Is., Solomon Islands. Other species originally placed in Ideohlothrus by their authors have been disposed of as follows: Ideobisium (Ideohlothrus) strandi Ellingsen 1901, assigned to Microcreagris Chamber- lin (Beier 1932:155), more recently to Syarinus Chamberlin (Mahnert 1976:206). /. roszkovskii Redikorzev 1922, designated the type species of Pseudoblothrus Beier (1931:21). Species originally assigned to Ideobisium (Ideoroncus) are not considered here. They have all proved to be referrable to various genera in the family Ideoroncidae, or in one case to Syarinus , family Syarinidae. DISCUSSION On the basis of the studies reported above, our concepts of the families Syarinidae and Neobisiidae must be changed in some important respects. Representatives of the Syarinidae share many characters with those of the Neobisiidae and clearly belong the superfamily Neobisioidea (Chamberlin 1930; Beier 1932, 1961, 1963; Hoff 1956, 1964). They have always been separated from the Neobisiidae by i) the nature of the abdominal pleural membranes — longitudinally striate in Syarinidae but granulate in Neobisiidae, and ii) the suture between the parts of the femur of leg IV - oblique to the long axis in Syarinidae but perpendicular in Neobisiidae (Chamberlin 1930; Hoff 1958; Beier 1963). However, difficulties with these distinctions have long been recognized. As already pointed out by Mahnert (1979:750) the pleural membranes are somewhat varied between granular and striate in species of Ideobisium , and especially so in Alocohisium Beier (now considered a genus of the Syarinidae, according to Mahnert 1974:851). In addition, it can be noted that Morikawa (1963) states explicitly that the pleural membranes of the abdomen of A. solomonense are striated, not granulated. Chamberlin (1938), describing the femoral suture of Pachychitra as only “weakly oblique 218 THE JOURNAL OF ARACHNOLOGY to vertical,” nevertheless assigned the genus to the Syarinidae. Hummelinck (1948) described the suture of P. curazavia as vertical; for Alocobisium Beier (1952) reported the articulation line “vertical to the long axis of the femur”; and the femoral sutures of Microcreagrella, Micro creagrina and Hadoblothrus are characterized by Beier (1963) as “senkrecht zur Langsaches des Gliedes,” It is evident, therefore, that the pleural mem- branes and the femoral suture of leg IV are not uniform throughout the Syarinidae and frequently are similar to those found in the Neobisiidae; they cannot then be used as key characters for distinguishing members of the two families. The taxa placed in the Syarinidae form a very diverse group. Three subfamilies have been recognized: Syarininae, Chitrellinae, and Microcreagrellinae. However, Ideobisium and Ideoblothms do not fit comfortably into any of these and the positions of Nan- nobisium and Alocobisium have not been determined with certainty. Nevertheless, there seem to be two characters which can be used to separate (most of) the Syarinidae from the Neobisiidae; these are: i) The apex of the palpal coxa is usually more or less triangular and bears two setae (rather than three or more). All the genera in question show this character clearly except Syarinus. In the latter, the apex of the coxa is low and rounded, so that the two apical setae are set very close to the setae bordering the trochanteral fossa and may be difficult to distinguish from them. ii) Trichobothrium t (on the movable finger of the palpal chela) is often shorter than the other trichobothria and is flattened, or lanceolate, toward the distal end. As reported previously (Muchmore 1979), this feature has been seen by me in representatives of Syarinus, Chitrella, Ideobisium, Ideoblothrus (syn. Pachychitra), and Nannobisium (syn. Vescichitra). Mahnert (in litt.) has reported finding lanceolate trichobothria t in the West African species Nannobisium liberiense Beier and Ideoblothrus lepesmei (Vachon). In two specimens of Alocobisium solomonense Morikawa (det. M. Beier) examined by me, t is shorter than the others and appears slightly flattened near the end. On the other hand, in two paratypes of Ideoblothms roszkovskii Redikorzev (type species of Pseudoblothms Beier), trichobothrium t is not different from the other trichobothria. And Mahnert (in litt.) reports on the situation in some specimens in the Museum d’Histoire Naturelle, Geneve, as follows: Micro creagrina hispanica: t only slightly shorter, not lanceolate Microcreagrella c. caeca: t shorter, not lanceolate Troglobisium racovitzai: t nearly the same length, not lanceolate Pseudoblothms ellingseni: t nearly same length, not lanceolate Hadoblothms gigas: t not shorter, not lanceolate It appears, therefore, that in the forms from the Americas, the Pacific area, and Africa south of the Sahara, trichobothrium t is characteristically shortened and flattened, while in those from Mediterranean Europe and North Africa, t may or may not be shortened, and is not flattened. Members of the family Syarinidae can, therefore, be recognized as follows: with the characters of the superfamily Neobisioidea, that is, all legs diplotarsate and chelicera with distinct teeth on both fingers; venom apparatus present only in fixed finger of palpal chela; apex of palpal coxa with two setae; and chela with usual complement of 12 trichobothria, of which t is usually short and lanceolate (except in European and North African forms). Within the Syarinidae pleural membranes may range from striate to granulate; the suture between the parts of the femur of leg IV may be oblique or MUCHMORE-THE GENERA IDEOBISIUM AND IDEOBLOTHR US 219 perpendicular to the long axis of the femur; a galea may be present or not; if present, the galea may be simple or divided; all trichobothria may be confined to the fingers or some may be located on the chelal hand; subterminal tarsal setae may be simple or denticulate; eyes may be present or absent. The following couplet may now be used to distinguish between Syarinidae and Neobisiidae: Apex of palpal coxa usually triangular and with two setae; trichobothrium t short and lanceolate (except in European and North African forms); pleural mem- branes granulate or striate; if pleural membranes granulate, the chelal hand bears one or more trichobothria dorsally or laterally) Syarinidae Apex of palpal coxa rounded and with three or more setae; trichobothrium t similar in form to other trichobothria; pleural membranes granulate; trichobothria confined to chelal fingers Neobisiidae The genera which comprise the family Syarinidae are Syarinus Chamberlin 1925 Chitrella Beier 1932 Aglaochitra Chamberlin 1952 Pseudohlothrus Beier 1931 Troglobisium Beier 1939 Hadoblothrus Beier 1952 Ideobisium Balzan 1891 Ideoblothms Balzan 1891 (syn. Pachychitra Chamberlin 1938) Nannobisium Beier 1931 (syn. Vescichitra Hoff 1964) Microcreagrella Beier 1961 Microcreagrina Beier 1961 Alocobisium Beier 1952 The remarkable species Hyarinus hesperus Chamberlin (1925) may belong here as suggested by Chamberlin (1930). However, Chamberlin’s original description is not sufficiently detailed to allow a definite decision and the holotype, the only known specimen, cannot now be located. ACKNOWLEDGMENTS I am greatly indebted to the following curators for allowing me to examine material in their care: L. F. deArmas [ACC] , E. M. Benedict [JCC] , J. Heurtault [MNHN] , H. W. Levi [MCZ] , N. I. Platnick [AMNH] , and F. R. Wanless [BM(NH)] . And I am grateful to the following collectors who contributed specimens: N. P. Ashmole, P. Chapman, H. J. Harlan, J. A. Kochalka, W. E. McMahon, and S. B. and J. Peck. I acknowledge with gratitude the information on European forms and critical comments on parts of the manuscript provided by V. Mahnert. C. H. Alteri prepared most of the illustrations. LITERATURE CITED Balzan, L. 1891. Voyage de M. E. Simon au Venezuela: Chernetes (Pseudoscorpiones). Ann. Soc. Ent. France, 60:497-552. Banks, N. 1909. New Pseudoscorpionida. Canadian Entomol., 41:303-307. 220 THE JOURNAL OF ARACHNOLOGY Beier, M. 1930. Die Pseudoskorpione des Wiener Naturhistorischen Museums. III. Ann. Naturhist. Mus. Wien, 44:199-222. Beier, M. 1931. Neue Pseudoscorpione der U. O. Neobisiinea. Mitt. Zool. Mus. Berlin, 17:299-318. Beier, M. 1932. Pseudoscorpionidea I. Subord. Chthoniinea et Neobisiinea. Das Tierreich, 57:1-258. Beier, M. 1939. Die Pseudoscorpioniden-Fauna der iberischen Halbinsel. Zool. Jahrb. Syst., 72:157-202. Beier, M. 1940. Die Pseudo scorpionidenfauna der landfernen Inseln. Zool. Jahrb. Syst., 74:161-192. Beier, M. 1947. Zur Kenntnis der Pseudoscorpionidenfauna des siidlichen Afrika, insbesondere der siidwest- und sudafrikanischen Trockengebiete. Eos (Madrid), 23:285-399. Beier, M. 1948. Uber pseudoscorpione der australischen Region. Eos (Madrid), 24:525-562. Beier, M. 1952. On some Pseudoscorpions from Malaya and Borneo. Bull. Raffles Mus., 24:96-108. Beier, M. 1954. Pseudoscorpionidea. Beitr. Fauna Perus, 4:1-12. Beier, M. 1955. Pseudoscorpionidea, gesammelt wahrend der schwedischen Expeditionen nach Ostafrika 1937-38 und 1948. Ark. Zool., 7:527-558. Beier, M. 1957. Pseudoscorpionida. Insects of Micronesia, 3(1 ) : 1-64. Beier, M. 1959. Pseudoscorpione aus dem Belgischen Congo gesammelt von Herrn N. Leleup. Ann. Mus. Roy. Congo Beige, Tervuren, 72:1-69. Beier, M. 1961. Pseudoscoprione von den Azoren und Madeira. Bol. Mus. Mun. Funchal, 14:67-74. Beier, M. 1963. Ordnung Pseudoscorpionidea. Bestimmungsbiicher zur Bodenfauna Europas, 1:1-313. Beier, M. 1964. Further records of Pseudoscorpionidea from the Solomon Islands. Pacific Insects, 6:592-598. Beier, M. 1965. Die Pseudoscorpioniden Neu-Guineas und der benachbarten Inseln. Pacific Insects, 7:749-796. Beier, M. 1966. Die Pseudoscorpioniden der Solomon-Inseln. Ann. Naturhistor. Mus. Wien, 69:133-159. Beier, M. 1967. Die Pseudoscorpione der Noona Dan Expedition nach den Philippinen und Bismarck Inseln. Entomol. Medd., 35:315-324. Beier, M. 1968. Some cave-dwelling Pseudoscorpionidea from Australia and New Caledonia. Rec. S. Australian Mus., 15:757-765. Beier, M. 1970. Erganzungen zur Pseudoskorpionidenfauna der Kanaren. Ann. Naturhist. Mus. Wien, 74:45-79. Beier, M. 1974. Ein neuer Par alio chthonius aus Guatemala. Rev. Suisse Zool., 81:101-102. Beier, M. 1976. Pseudoscorpione von der Dominicanischen Republik (Insel Haiti). Rev. Suisse Zool., 83:45-58. Chamberlin, J. C. 1925. On a collection of pseudoscorpions from the stomach contents of toads. Univ. California Publ., Tech. Bull. Coll. Agric., Entomol., 3:327-332. Chamberlin, J. C. 1930. A synoptic classification of the false scorpions or chela-spinners, with a report on a cosmopolitan collection of the same. Part II.-The Diplosphyronida (Arachnida-Chelonethida). Ann. Mag. Nat. Hist., (10)5:1-48. Chamberlin, J. C. 1931. The arachnid order Chelonethida. Stanford Univ. Publ. Biol. Sci., 7(1) : 1-284. Chamberlin, J. C. 1938. A new genus and three new species of false scorpions from Yucatan caves (Arachnida-Chelonethida). Publ. Carnegie Inst. Washington, 491:109-121. Daday, E. 1897. Pseudoscorpiones e Nova-Guinea. Termesz. Fiizetek, 20:475-480. Ellingsen, E. 1901 . Sur une espece nouvelle d'Ideobisium genre des Pseudoscorpions de l’Europe. Bull. Soc. Zool. France, 26:86-89. Ellingsen, E. 1909a. On some North American pseudoscorpions collected by Dr. F. Silvestri. Boll. Lab. Zool. Portici, 3:216-221. Ellingsen, E. 1909b. Contributions to the knowledge of the pseudoscorpions from material belonging the the Museo Civico in Genova. Ann. Mus. Civ. Stor. Nat. Genova, 4:205-220. Ellingsen, E. 1910. Die Pseudoskorpione des Berliner Museums. Mitt. Zool. Mus. Berlin, 4:355-423. Ellingsen, E. 1912a. Pseudoscorpions from Formosa. Nytt Mag. Naturvid., 50:121-128. Ellingsen, E. 1912b. Biospeologica 26. Pseudoscorpiones (3 ser). Arch. Zool. Exp. Gen., (5) 10:163-175. Ellingsen, E. 1912c. The pseudoscorpions of South Africa based on the collections of the South African Museum, Cape Town. Ann. S. African Mus., 10:75-128. Hoff, C. C. 1945. New neotropical Diplosphyronida (Chelonethida). Amer. Mus. Novitates, 1288:1-17. Hoff, C. C. 1956. Diplosphyronid pseudoscorpions from New Mexico. Amer. Mus. Novitates, 1780:1-49. MUCHMORE-THE GENERA IDEOBISIUM AND IDEOBLOTHRUS 221 Hoff, C. C. 1958. List of the Pseudoscorpions of North America north of Mexico. Amer. Mus. Novitates, 1875:1-50. Hoff, C. C. 1964. The pseudoscorpions of Jamaica. Part 3. The suborder Diplosphyronida. Bull. Inst. Jamaica, Sci. Ser., 10(3): 1-47. Hummelinck, P. W. 1948. Pseudoscorpions of the genera Garypus, Pseudochthonius, Tryannoch- thonius and Pachychitra. Stud. Fauna Curasao, Aruba, Bonaire, Venezuelan Is. 3(1 3) :29-77. Mahnert, V. 1974. A can thocreagris nov. gen. mit Bemerkungen zur Gattung Microcreagris (Pseudo- scorpiones, Neobisiidae) (Uber grieschische Pseudoskorpione IV). Rev. Suisse Zook, 81:845-885. Mahnert, V. 1976. Zur Kenntnis der Gattungen Acanthocreagris und Roncocreagris (Arachnida, Pseudo scorpio nes, Neobisiidae). Rev. Suisse Zook, 83:193-214. Mahnert, V. 1978a. Pseudoscorpione (ausgenommen Olpiidae, Garypidae) aus Congo-Brazzaville (Arachnida, Pseudoscorpiones). Fol. Entomol. Hungarica, 31:69-133. Mahnert, V. 1978b. Contributions a l’etude de la faune terrestre des lies granitiques de l’archipeldes Sechelles (Mission P. L. G. Benoit-J. J. Van Mol 1972), Pseudoscorpiones. Rev. Zook Afr., 92:867-888. Mahnert, V. 1979. Pseudoscorpione (Arachnida) aud dem Amazonas-Gebiet (Brasilien). Rev. Suisse Zook, 86:719-810. Mello-Leitao, M. 1937. Novo pseudoscorpiao do Brasil. Ann. Acad. Brasileira Sci., 9:269-270. Morikawa, K. 1963. Pseudoscorpions from Solomon and New Britain. Bull. Osaka Mus. Nat. Hist., 16:1-8, p. 1-4. Muchmore, W. B. 1972. New diplosphyronid pseudoscorpions, mainly cavernicolous, from Mexico (Arachnida, Pseudoscorpionida). Trans. Amer. Micros. Soc., 91:261-276. Muchmore, W. B. 1979. Pseudo scorpions from Florida and the Caribbean area. 7. Floridian diplosphy- ronids. Florida Entomol., 62:193-213. Murthy, V. A. and T. N. . Ananthakrishnan. 1977. Indian Chelonethi. Oriental Insects Monogr., 4:1-210. Redikorzev, V. 1922. Pseudoscorpions nouveaux I. Akad. Nauk, S.S.S.R. Zook Muz., 22:91-101. Simon, E. 1 880. Materiaux pour servir a une faune Arachnologique de la Nouvelle Caledonie. C. R. Soc. Entomol. Belg., 23:164-175. Stecker, A. 1875. Uber neue indische Chernetiden. Sitz. math.-phys. Kl. bayer. Akad. Wiss., 72:512-526. Tullgren, A. 1905. Einige Chelonethiden aus Java. Mitt, naturh. Mus. Hamburg, 22:37-47. Tullgren, A. 1907. Chelonethiden aus Natal und Zululand. Zook Stud, tillagn. Tullberg: 2 16-236. Vachon, M. 1941. Pseudoscorpions recoltes en Afrique occidentale tropicale par P. Lepesme, R. Paulian et A. Villiers (Note preliminaire). Bulk Sci. Bourgogne (Dijon), 9:29-35. With, C. J. 1905. On Chelonethi, chiefly from the Australian region, in the collection of the British Museum, with observations on the “coxal sac” and on some cases of abnormal segmentation. Ann. Mag. Nat. Hist., (7) 15:94-143. With, C. J. 1906. An account of the Indian false-scorpions together with studies on the anatomy and classification of the order. Kgl. Danske Vidensk. Selsk. Skrift., (7) 3:1-214. Manuscript received May 1981, revised September 1981. Francke, O. F. and S. K. Jones. 1982. The life history of Centruroides gracilis (Scorpiones, Buthidae). J. Arachnol., 10:223=239. THE LIFE HISTORY OF CENTRUROIDES GRACILIS (SCORPIONES, BUTHIDAE) Oscar F. Francke and Steven K. Jones Departments of Biological Sciences and Entomology, and The Museum, Texas Tech University Lubbock, Texas 79409 ABSTRACT Laboratory-reared females mature at the seventh instar, which is reached at the age of 302.5 ± 25.2 days (n = 24). Males mature at either the sixth or the seventh instar and experience no postmaturation molts. “Small” males reach adulthood at the age of 235.7 ± 14.8 days (n = 14), whereas “large” males require 281.8 ± 10.7 days (n = 19). Large males are approximately 1.25 times bigger than small males in the lengths of six structures measured (carapace, femur, tibia, chela, metasomal segment V, and telson), and produce spermatophores that differ by the same ratio from those of small males. Growth rates are discussed with respect to sexual dimorphism, and the instar at which maturity is reached. Attaining maturity at different instars appears to be a common life history strategy among buthid scorpions: in some species males mature at different instars, in others females do, and in still others both sexes are variable. INTRODUCTION Buthids are the most numerous and widely distributed (worldwide) family of Recent scorpions, representing about 45 of 115 genera (39%) and 600 of 1200 described species (50%). All scorpions possess venom glands, but the dozen or so species considered danger- ous to mammals, including man, are buthids. Consequently there has been greater interest and more research done on buthids than on any other scorpion family. Studies on scorpion life histories, however, are few in number. Parameters such as litter size, number of molts, and age to maturity are known only for about 20 species, half of which are buthids. The lack of more data relates in part to the difficulties met in rearing scorpions in captivity (Francke 1976, 1979a, 1981, Polis and Farley 1979). In North America buthids are represented by the genus Centruroides Marx, of which at least six species are considered medically important (Keegan 1980). This is the first complete life history study of any North American buthid. In the present study 52 of 72 Centruroides gracilis (Latreille), of two litters born in captivity, were raised to sexual maturity (success rate of 72%), and the average age to maturity was less than 300 days (range 214 to 348). The adaptability of this species to laboratory conditions and its rapid rate of development make it an excellent subject for studies of ontogenetic changes and variability. In turn, this aids understanding of many other poorly known aspects of scorpion biology. 224 THE JOURNAL OF ARACHNOLOGY The objectives of this paper include detailed considerations of the life history of C gracilis: e. g., number of molts, chronology, age to maturity, growth rates and allometry, sexual dimorphism, ontogenetic variability in pectines and trichobothria, and sperma- tophore differences between males which mature at different instars. MATERIALS AND METHODS Two adult female Centruroides gracilis were collected 13 km E of Xilitla (500 m elevation), San Luis Potosi, Mexico, on 10 March 1977. They were returned alive to Lubbock, Texas, where on 16 May 1977 each gave birth to litters of 26 and 46 young, respectively. The young positioned themselves randomly over their mother’s dorsum and underwent their first molt eight days later. They dispersed on 28-31 May, at which time they were sorted into individual containers. Young scorpions were kept in 75 ml wide- mouth jars (50 mm internal diameter), with a semicircular piece of paper towel on one side and a small piece of moistened sponge on the other. Upon reaching the fourth instar the specimens were transferred to 11x11x7 cm plastic containers lined with paper towels and provided with a small watch glass filled with water. Specimens were kept in an environmental chamber at 26.6 ± 2°C. Darkness was inter- rupted only during maintenance activities, which occurred at various hours of the day. The specimens were checked and watered daily, at which time any molts or deaths were noted and recorded. Prey was presented on alternate days and consisted mainly of live immature cockroaches, Nauphoeta cinerea (Saussure). Pectinal tooth counts and measurements of the length of six structures (carapace; pedipalp femur, tibia, and chela; metasoma segment V, and telson) were obtained from each exuvium, preserved specimen, or live adult, to analyze both variability and the growth factor per molt (Dyar’s “constant”). Centruroides gracilis males attain sexual maturity at two different instars. Conse- quently, statistical analyses were initially performed on a 3 x 2 (“sexes” X litters) fac- torial analysis of variance. The means were compared by Duncan’s multiple range test (Steel and Torrie 1960). Many deaths were associated with molting. For individuals dying during molts, the duration of the previous instar was recorded but morphometric data for the succeeding instar could not be obtained. Therefore, differences in the number of individuals reported in different sections of the text reflect properties of the data sets. Data on buthid life histories were obtained from the literature, and where possible pertinent parameters were calculated from available raw data. Observations on litter size in Centruroides were obtained from the literature, and from preserved museum specimens in various collections. RESULTS AND DISCUSSION This section is divided into four main parts, each dealing with a specific aspect of life history phenomena in C. gracilis: 1) litter size; 2) basic life history parameters such as number of molts, age to maturity, and survivorship; 3) growth rates and allometry; and 4) ontogenetic variability in various morphological characters. Throughout the rest of this paper males attaining sexual maturity upon reaching the sixth instar will be called “small males”; those attaining sexual maturity upon reaching the seventh instar will be “large males.” FRANCKE AND JONES-LIFE HISTORY OF CENTRUROIDES GRACILIS 225 Table 1. -Female size (carapace length used as a first order approximation) and litter size in the genus Centruroides ; carapace length for C. insulanus from Thorell (1876) and Pocock (1893), litter size from Baerg (1954). Carapace length female Litter size n x ± s.d. n x ± s.d. C. exilicauda (Banks) 8 5.3 ±0.3 8 24.0 ± 8.6 C. gracilis (Latrielle) 5 8.7 ± 0.6 6 42.5 ± 25.7 C. griseus (Koch) 1 6.1 1 35 C. insulanus (Thorell) 2 6.4 ±0.1 11 50.0 ± 33.0 C. margaritatus (Gervais) 6 8.3 ± 0.4 6 39.7 ± 15.3 C. vittatus (Say) 10 5.0 ± 0.3 10 22.6 ±5.8 Litter size— The number of young per litter in C. gracilis is quite variable: Lucas (1890) reported a litter of 91 young from Panama, and Armas (1980) litters of 22 and 34 young from Cuba. In addition to the litters of 26 and 46 young from Mexico which are the subject of this paper we have examined females with their litters from Florida (n = 30 young), Belize (n = 42 young), and Honduras (n = 20 young). Intraspecific variability in litter size in scorpions has not been studied. Francke (1981), however, showed that 81% of the variability (both intra- and interspecific) in litter size among diplocentrid scorpions was accounted for by differences in female size and size of young at birth. Variability in litter size in Centruroides spp. is summaried in Table 1. Intraspecific variability in the data is sufficiently large (coefficients of variation range from 25.6% to 66.0%) so as to render statistically meaningless any analyses on inter- specific variability among Centruroides spp. at this time. Although it is possible that the data in Table 1 reflect natural variation, in some instances (particularly preserved museum samples) other factors such as maternal cannibalism, dispersal of part of the Utter prior to capture, and careless preservation of the young could seriously affect the results. It is hoped, however, that as more and better data become available it will be possible to determine the factors affecting litter size in the genus Centruroides , and in buthid scor- pions in general. Life history.— Litter I consisted of 26 individuals of which 20 reached sexual maturity: 2 small males, 8 large males, and 10 females. Five specimens died in the second instar, two of unknown causes and three of complications associated with the molt to third instar. The sixth and last death occurred during the molt from fourth to fifth instar. Litter II consisted of 46 individuals of which 32 reached sexual maturity: 12 small males, 11 large males, and 9 females. One young died of unknown causes during the second instar, three died molting to the third instar, one died during the fifth instar, three died during the molt to sixth instar, and six died during (or shortly after) the molt to seventh instar. Therefore, the largest source of mortality in this part of the study was molting (80% of the 20 deaths). The first instar lasted eight days in both litters. Statistics on the duration (in days) of each succeeding instar of C. gracilis are summarized in Table 2. The duration of the second instar was significantly different between litters I and II (F = 7.56, d.f. = 1, p = 0.008); we are unable to determine the biological meaning, if any, of this difference. The duration of the third instar was similar in both litters, suggesting that factors responsible for the difference between the duration of the second instar were temporary. However, 226 THE JOURNAL OF ARACHNOLOGY the duration of the fourth instar was again statistically significantly different between litters (F = 41.05, d.f. = 1, p = 0.0001), with litter II requiring considerably longer than litter I (this is the reverse of the pattern observed on the second instar). Large males require significantly fewer days than small males and females to complete the fifth instar (F = 10.51 , d.f. = 2, p = 0.0002). Small males are sexually mature upon reaching the sixth instar. The average age to maturity in small males is 235.7 ± 14.8 days. There are no significant differences in the duration of the sixth instar for large males and females. There are, however, significant differences in the total age to maturity: large males mature at 281.1 ± 10.7 days of age, and females mature at 302.5 ± 25.2 days. The earliest maturing specimen, a small male, reached the sixth and final instar at 214 days of age; the latest maturing specimen, a female, reached the seventh instar at 348 days. The data suggest a continuous trend for slightly faster development (i. e., shorter intervals between molts) in large males with respect to small males and females; by the fifth instar the differences become statistically significant. Although the sixth instar of large males lasts 65.2 ± 13.4 days, they reach the seventh instar (and sexual maturity) only 46.1 days after small males do. Mating between a large male and a female was attempted on 26 February 1978, approximately two weeks after their respective final molts. The male engaged in courtship behavior but the female was unresponsive. Three more matings with large males were attempted on 2 April 1978, and they were also unsuccessful. Six matings were attempted on 23-29 April 1978, two with small males and four with large males. Three spermato- phores in the preinsemination condition (Francke 1979b) were recovered: one from a small male and two from large males. Four additional matings were attempted on 6-7 February 1979, with three small and one large male. Two spermatophores in the post- insemination condition were recovered, one from the large male and one from a small male. Finally, five matings were attempted in Febraury-March 1980 with two small and three large males; no spermatophores were recovered. Therefore, sexual maturity in both small and large males has been confirmed by staged matings in the laboratory during which spermatophores were produced. Postmaturation molts, i. e., a small male molting into a large male, did not occur in over 3 years of continued observation. Fig. 1.- Survivorship (%) as a function of time in two litters of Centruroides gracilis born and raised in the laboratory. Hatched region (February-April 1978) represents the final molt (to sexual matur- ity). Solid circles = litter I (n = 26); open circles = litter II (n = 46). FRANCKE AND JONES- LIFE HISTORY OF CENTRUROIDES GRACILIS 227 r X dJ o £ o X o 0 1 .s X . -4™* /""N ^ to X « 00 ^ 3 Oi O w> 5 s S3 3 a M .£ +' x IX s on q on +i IX 3 o 52 00 +l IX q oo +1 IX q oo +1 IX o 0) o t-i M H H 3 3 3 3 3 cd a cd cd a cd £ £ £ £ £ £ 00 o d- OS d- i— i '—l CM X X cd X cd cd cd Os CM X p d; wo cd vd rH cd 1-H vd CM .-h r— H r-H CM 2 +1 +1 as OS 3 CM X «3 © CD cd 00 wd 3 wd CM £ VO 00 £ X X £ X c-~ CM 00 O CM rH d- ■d Os d- r-H r-H 1 rH CM ■8 X ab cd X cd cd X cd CM © d; p Os wo vd X © in CM CM wo d" wo ^H OS r-H rH as 00 r-H +1 +1 +1 +l +1 +i +i +1 +1 wo wo cq d; P p X Cq r-H rH CO d c- cd cd vd cd wo VO wo X X X X X wo X CM 00 o CM r— 1 X d- os X i— ( ,""i 1 rH H 1 CM ab cd cd X X X 3 cd cd d; cq CM r-H p Os CM p wo o wo vd cd X d" as cd cd r-H +i +i +i +1 +l +i +1 +1 +1 © OS CM cq CM os CO p CM d- © d r-H d- © vd Os wo wo wo X X X X wo in CM 00 © CM X d- Os X r-H r*H rH CM 03 cd cd cd cd cd cd cd cd "d oq cq Cq p cq rq Cq d; VO WO cd wo wo d1 wd wd vd +1 +i +i +l +1 +l +i +1 +i wo p CM cq CM rH CM p CM On CD Os CM as d" CM OS CM d" WO d" in d* in WO ■d- X CM 00 © CM X d- Os X r™H 1—1 i— i H r-H CM o o X X X o c3 cd cd cd a X cd cd cd CM Os wo Os d; CM Cq OS Cq ON CM ■d vd cd vd cd vd wd +1 +1 +l +i +1 +i +1 +i +1 wo wo r— 1 X 1—1 in © CM wo as CM © cd wo cd r-H 00 oo W0 wo wo d- d- d- wo d- d- CM OO © CM X "d- Os X 1—1 1 ' 1 r-H t-H CM <*> C/3 i/i C/3 0> 1— 1 £ M cd + £ cd 0) on X [X oo X X M 00 X X M 4) — i CM cq vo CM CO +1 +1 Os Cq r~- co wo vo o os CM CO cq in d- cd +i +i oo cq O CM* in vo O OS CM CO vo VO vo in +i +i oo co On CM d- vo O ON CM CO © 00 wd VO +1 +1 © wq cm cd wo d- O OS CM CO 4) O ALL 59 49.0 ± 6.6 59 51.4 ± 6.0 59 58.7 ± 8.7 59 61.8 ± 13.4 41 69.2 ± 18.4 228 THE JOURNAL OF ARACHNOLOGY Laboratory reared females are slightly smaller than their mothers, but the size dif- ferences are not large enough to postulate an additional molt before the attainment of sexual maturity (Table 3). The two females involved in the matings which yielded postinsemination spermatophores failed to produce any young in over 1.5 years of observation. However, two unmated females at the seventh instar aborted what appear to be mature oocytes or very early embryos (approx. 2 mm in diameter), indicating that those females are indeed sexually mature (Parthenogenesis is known to occur in other buthid scorpions; Matthiesen 1962). Mortaility among immatures was not significantly different between litters as approxi- mately 70% of the individuals born in the laboratory reached sexual maturity (Fig. 1). As indicated ear her, most deaths were associated with molting. Adults live about two years in the laboratory, at which time senescence apparently occurs (Fig. 2). Individuals from litter I lived on the average six months longer than individuals from litter II; however, we cannot explain this difference between litters and perhaps it is spurious. Among speci- mens that reached sexual maturity there are no differences in survivorship between small males and large males (Fig. 2), each lives an average of 33 months in the laboratory. Females have an average life expectancy of 38 months. The longest lived small male died on 20 July 1981, the longest lived large male was also the longest lived individual and died on 7 September 1981, and the longest lived female died on 6 August 1981 . Growth and allometry.— First instar scorpions have a poorly sclerotized exoskeleton. Upon molting a very thin, fragile, considerably wrinkled and folded exuvium is recover- able. However, it is not possible to obtain accurate measurements of any body parts from these exuvia. Thus, the following information on growth and allometry is restricted to the second through seventh instars. Second instar specimens from litter I are significantly smaller than those from litter II (Tables 4 and 5). Since first instars do not feed but just complete development, the size differences noted are most probably a reflection of differences in size at birth. On all structures measured, the growth factors associated with the molts from second to third, third to fourth, and fourth to fifth instars are not significantly different between litters (Table 5). Consequently, the relative differences in size (on all structures) between litters Fig. 2.- Survivorship (%) as a function of time in Centruroides gracilis individuals (from two litters) which attained sexual maturity in the laboratory. Open circles = males that matured at the sixth instar or “small males” (n = 14); solid circles = males that matured at the seventh instar or “large males” (n = 17); triangles = females (n = 21). FRANCKE AND JONES-LIFE HISTORY OF CENTRUROIDES GRACILIS 229 Table 3. -Comparisons of sizes of various structures between mothers and their laboratory reared daughters (measurements in mm). Predicted eighth instar dimensions for daughters were determined by mutliplying the mean size at the seventh instar times the growth factor of that structure during the previous molt (see Francke 1976, 1979a, 1981 for details). Females Daughters 7 th Predicted 8th I II mean min. max. Carapace length 8.85 9.00 8.09 7.80 8.55 10.35 Femur length 8.65 8.70 7.84 7.35 8.55 10.37 Tibia length 9.15 9.60 8.50 8.10 9.15 11.19 Chela length 15.00 15.45 13.75 13.05 14.70 18.12 Segment V length 11.70 11.70 9.34 8.40 10.65 12.79 Telson length 8.10 8.40 7.12 6.60 8.20 9.29 at birth remain proportionally unchanged through the various molts (see Table 4 for details on carapace length and growth factors). The genus Centruroides Marx is characterized by the strong sexual dimorphism in metasoma length in adults (Stahnke and Calos 1977). In addition, various morphometric ratios have been proposed for the identification of species (Stahnke and Calos 1977). Despite the increasingly important role that morphometries are assuming in the tax- onomy of the genus, there have been no detailed studies of variability in the characters. Therefore, we analyzed growth parameters on six structures of C. gracilis. The length of the carapace at each instar, and the growth factor associated with each molt, for the six sex-litter groups appear on Table 4. Note that there are no significant differences due to sex in carapace between second instars, whereas there are significant differences between adults. The results of comparable analyses done on the other five structures are sum- marized in Table 5. Sexual dimorphism in metasomal segment V length is expressed as early as the second instar in C. gracilis , and develops gradually in the other five structures. The trends in allometry are given in Figs. 3-8, and are briefly discussed below for each structure because of their significance to any future attempts to use morphometries in the taxonomy of the genus. CARAPACE. There are no significant differences in growth rate of carapace length (CL) between the two litters during the second through fifth molts (Table 5). During the sixth and final molt for large males and females there is a significant difference: the CL growth factor on litter I was 1.29 ± 0.06, and on litter II it was 1.23 ± 0.04. Analysis of variance by Utter X sex cohorts shows that females from litter I had a higher CL growth rate than either males or Utter II females (Table 4). The rate of CL growth is not significantly different between sexes during the second through fourth molts (Table 5, Fig. 3). During the fifth molt small males, large males, and females experience a significant and unequal reduction in the rate of CL growth, with small males experiencing the largest decrease in this their last molt (Fig. 3). Large males experience a signficant reduction in CL growth rate during their sixth (and final) molt, whereas females show no significant differences at the sixth molt. There are no significant differences in either CL or CL growth rate due to sex X Utter interactions. However, the initial (second instar) differences in CL’s between Utters, com- pounded with the differences in growth rates due to sexual dimorphism and all°metry on the fifth and sixth molts, results in six significantly different clusters of CL’s among adults (Table 4). 230 THE JOURNAL OF ARACHNOLOGY FEMUR. There are no significant differences due to sex in femur length (FL) in second instars, nor in the growth rate between litters during the second through fifth molts (Table 5). The two litters differ significantly in FL growth rates during the sixth and final molt for females and large males, with litter I experiencing a higher FL growth rate than litter II. There is, however, a significant litter X sex interaction (Table 5) and Duncan’s multiple range test indicates that litter II females grew significantly less than either males or litter I females during that molt. Sex is not a significant factor in differences in FL growth rates until the fifth molt. Concurrent with this molt small males grew at a significantly higher rate than large males and females, and also at a higher rate than in the previous three molts (Fig. 4). The FL growth rate on large males remains approximately the same during the second through fifth molts, whereas the FL growth rate on females decreases significantly with the fifth molt. Femur length at the sixth instar is 7.07 ± 0.30 mm (n = 13) for small males (mature), 6.43 ± 0.20 mm (n = 19) for large males (subadult), and 5.99 ± 0.36 mm (n = 24) for females (subadult). At the seventh instar it is 9.04 ± 0.29 mm (n = 19) for large adult males, and 7.84 ± 0.31 mm (n = 20) for adult females. TIBIA. There are no significant differences in the growth rate of tibia length (TL) during the second and third molts. On the fourth molt there are significant differences between sexes, with small males experiencing a proportionately greater increase in TL than do females and large males. Associated with the fifth molt are significant differences due to sex and due to sex X litter interaction: females have a lower TL growth factor than males, and than they had in previous molts; and both large and small males from different litters are significantly different from each other (Table 5). The growth factors associated with the sixth molt are significantly different between litters, sexes, and due to interactions of these two factors. Duncan’s multiple range test indicates that litter II females had a lower TL growth rate than do either males or litter I females. The length of the tibia on sixth instars is 7.46 ± 0.39 mm in small males (adult), 6.95 ± 0.22 mm in large males (subadult), and 6.59 ± 0.41 mm in females. The tibia length in seventh instars is 9.49 ± 0.32 mm and 8.50 ± 0.30 mm for large males and females respectively. CHELA. The chela length (CHL) growth factor shows no significant differences until the fourth molt. At this point there are no significant differences due to sex or litter, but there are significant differences between small and large males. These dif- ferences also result in a significant litter X sex interaction (Table 5) which appears to be spurious. The fifth molt results in significant differences in CHL growth rates between sexes. The largest CHL growth rate occurs in small males, and the lowest in females. There are no significant differences between litters, nor is there a significant litter X sex interaction with this molt. The sixth molt indicates signifcant differences in CHL growth rates between males and females, and also between litter I and litter II females. Males experience greater elonga- tion of the chela than females; the CHL growth factor is larger than on previous molts, and similar in magnitude to that experienced by small males during their final molt. The CHL growth rate on females remains fairly constant through the various molts (Fig. 6). The length of the chela on sixth instars is 12.08 ± 0.65 mm in small adult males (n = 13), 1 1.05 ± 0.44 mm in large males (n = 19), and 10.59 ± 0.68 mm in females (n = 24). FRANCKE AND JONES-LIFE HISTORY OF CENTRUROIDES GRACILIS o C/3 cd g G cd rO o> ■G GI cd cd G v OO NO co ■'d- NO CO oo -d- oo co cd O o o i-H q q q q o q q £ £ o cd £ II d +i d +i o +i d +i o +i o +i d d +i +i d d +i +i CD Ft . c I/O CO NO CO CO q C/3 SG .CO ^CD C/3 /— \ Cd 1— 1 in m NO CO 1-H DO C- 00 C/3 Mh On CO q co NO CO q q 1-H CO 2 *c cd o G G d rH co CO t-H d i-H NO d O g G Q C“ \ wo £ q g Vi G cd cd cd G cd cd cd cd cd £ in "G in CO ON ^d- (N Cd i-H Cd G li 13 OO q O o O o q H O rH O w cd ,G £ II d d o d d d o d o d ,G M o< 3 > iG *— < G G x> 05 05 o t3 0> Vh s cd cd cd cd cd cd id e« cd £■1 o o cd Cd NO r- ■^f O ot- 'd- ■“ 1 O 3 g O o q o q d o rH G "g II d o d o d d o o O O •SP Gh cd 3 +i +i +i +i +i +i +1 +1 +1 3 £ o NO in -d" o wo o oo CO £ o DO i-H CO OO co 00 CO q q q a> G Cd r— 1 co’ i-H CO d wo ^ NO G B iG 53 ^ 3 Ml hG B ^ rG 5 1 >1 3 M , £ G , ^ % 53 M ^ S M ^ d G .£ G G o o .£ ,G C o G O G O G CD CD M g a •S 5 O »h d a ^ (D »H .£ d O 'rt £ G G tH 05 ,G H JO o o CD G X G O 4h s 1 00 H IG E cn 231 Seventh instar Length 8.26 ± 0.22 a 8.01 ± 0.18 c 8.34 ± 0.31 b 8.16 ± 0.22 d 232 THE JOURNAL OF ARACHNOLOGY l.itf - 1.42 - 1.40 - 1.24 - 1.22 - 1,20 - 1.18 J 1 | | M PI F 2nd holt m M F 3rd molt m M F 4th molt m PI F 5th molt 3 Pt F 6th molt 1.46 1.44 - 1.42- 1.28 - 1.26- 1.24 3 i—i—| 1 — | — | ! — | — , — m M F m M F m M F 2nd molt 3rd molt 4th molt m PI F 5th molt 4 M F 6th molt 1.46 " 1.44 - 1.42 - 1.40- t 1.38 - 1.28- 1.26 - 1.24 - 1.22 J 1-n 1— T— I — M H F m PI F 2nd molt 3rd molt m H F 5th molt 5 t 1,48 - 1.46 - 1.44 - 1.42 - 1.40 - 1.28- 1.26- 1.24-1 r_r_I 1 — | — , — M N F mUF mHF 2nd molt 3rd molt 4th molt TT~1 — mHF 5th molt 6 M F 6th molt 2nd molt 3rd molt 4th molt 5th molt 6th molt 2nd molt 3rd molt 4th molt 5th molt 6th molt Figs. 3-8.— Growth factors (GF = “Dyar’s constant”) per molt for small males (m), large males (M), and females (F) from two litters of Centruroides gracilis born and raised in the laboratory. Small males are sexually mature after the fifth molt (i. e., at the sixth instar) and cease to molt thereafter. Shown are the mean, ± one standard error of the mean (box), and ± one standard deviation. 3, Carapace length GF; 4, Pedipalp femur length GF; 5, Tibia length GF; 6, Chela length GF; 7, Metasomal segment V length GF; 8, Telson length GF. FRANCKE AND JONES-LIFE HISTORY OF CENTRUROIDES GRACILIS 233 On seventh instar adults chela lengths are 15.51 ± 0.64 mm in males (n = 19), and 13.75 ± 0.45 mm in females (n = 20). METASOMA SEGMENT V. The growth factor of metasomal segment V length (ML) shows no significant differences during the second and third molts (Table 5). Sexual differences in ML growth rates become significant with the fourth molt, and continue to be so through the fifth and sixth molts (Table 5). The pronounced allometry observed during the last molt of males (Fig. 7) doubtlessly accounts for a large portion of the well documented pattern of sexual dimorphism in the genus Centruroides : adult males have considerably longer metasomas than females. Significant sexual dimorphism, however, is apparent on second instars despite the highly significant differences in size between litters: ML on litter I second instar males is 1.96 ± 0.09 mm (n - 10), on females it is 1.89 ± 0.09 mm (n = 10); on litter II second instar males it is 2.27 ± 0.09 mm (n = 23), and on females it is 2.15 ± 0.08 (n = 16). Thus, within littermates there is an average difference of about 0.10 mm in ML between sexes. On the third instar the difference increases to about 0.15 mm, on the fourth instar to approximately 0.20-0.25 mm; on the fifth instar the difference is 0.70-0.80 mm between small males and females, 0.30-0.40 between large males and females, and 0.45-0.50 between large and small males with the latter having longer caudas. The differences between sixth instars are as follows: small males versus females 2.10-2.30 mm, small males versus large males 1.05-1.50 mm, and large males versus females 0.70-1.00 mm. Segment V lengths in adults are 9.30 ± 0.55 in small males, 12.08 ± 0.57 mm in large males, and 9.34 ± 0.62 in females. Note that although small males undergo one fewer molt than females, ML is similar in these two groups. TELSON. There are no significant differences in telson length (TEL) growth factors associated with the second, third, and fourth molts. During the fifth molt small males experience a higher rate of TEL increase than do either large males or females, and large males outgrow females. There is also a significant litter X sex interaction during the fifth molt, which appears to be spurious and due to the low TEL growth rate observed in litter I small males (n = 2). During the sixth molt large males again outgrow females, and there is also a significant difference between litters (Table 5). There are significant differences in TEL between litters from the second instar onward, with litter I being smaller than litter II, which is the pattern observed in all the structures studied (Table 5). Sexual differences in TEL appear on third instars and continue throughout. The trend in TEL differences parallels that observed in metasomal segment V length, i. e., small males > large males > females of a given instar, although it is not as pronounced in magnitude. TEL on sixth instars is 6.56 ± 0.43 mm in small (adult) males, 5.93 ± 0.31 mm in large (subadult) males, and 5.51 ± 0.36 in females. On seventh instars it is 8.12 ± 0.29 mm in males, and 7.12 ± 0.41 in females. Variability.— In addition to the morphometric variability already noted, there are other structures which for various reasons are considered in this section. PECTINAL TOOTH COUNTS. The number of teeth on each pectine in each in- dividual C. gracilis is fixed at birth and remains constant throughout the various molts. This observation could be of practical use in field studies in that pectinal tooth counts could be used as “through molts” tags in conjunction with other “between molt” marks to follow individuals in natural populations. (For example, in some site-tenacious species one of us has used various dots of flourescent paint as “between molts” markers for many individuals in a given population. However, when immatures molted the flourescent-paint marks were lost. If the pectinal tooth count of an individual is known, as is the location of its home range or burrow, then upon molting a slightly larger unpainted individual Table 5.— Morphometric analyses of growth during the life history of C. gracilis. A 3 X 2 (“sexes” X litters) factorial analysis of variance was performed on the length of each structure at each instar, and on the growth factor of each structure at each molt; values of F are given below (d.f. = degrees of freedom; levels of significance indicated are *=0.05, **=0.01). (a = small males mature, excluded from further analyses) 234 THE JOURNAL OF ARACHNOLOGY o in SO CN Os Os t-H so t"- CN oo so d- © q q o q q q q CO o q r-H CN q t-H T-H r-H iH d o d CO CN 1— 1 rH SO r- t-H d d 1 1—1 * * * * * * * * * * * * * * * * * * CN co o o as in as d- CO ^O in 00 CN so t-H q q q q CO t-H q q r-H q q q t-H q co co i CO CN in o CO CO d in co o d 00 d C" in d- co co CO * * * * * * * * in CN Os as o so i—i i -H r-~ Os 00 o so r- m q q q o q so t-H o Os 00 so t-H q os os d CN d o CN o o CN CN Os CN CN CN t-H d- o CN * * * * * * * * * * * * * * * * * * * Os CO o in so o 00 in T— | c-~ o- 00 00 m o q q in q q i—i o t-H o o CN q so CN q 00 »n d 00 d o SO 00 d in in d d 00 rH Os CO CN d- * * * * * * * * t-H in H t-H CO 00 c— Os r-H CO r- CN as CO CO CN CO o q q CO o q o q co q 00 as d t-H cn o d d o CO in o CO CN 00 co CO CO co * * * * * * * * * * * * * * * o o SO r~-~ co r-H CO m r-H d- OS co 00 q q q q q q q d- a s CN q q CN os CN r-H rH m o o 00 d d CO in d as d CN SO ■d- CO SO CN CN * * * * * * * * * * * * t-H CN CN t-H so CO os CN r- CN in in d- O d- SO CN q so q q CN q q o SO so q OS q d d CN CN CN rH CN in d d d «n co d CN T‘H t-H CN t— H * * * * * * * * * * * * * * * so in rH rH in oo CO d- CN in d- l-H t-H d- q o c — q q q o oo CO q q o q co co i-H d d co d r-H d CN d as in o t-H d so On in 00 CN ■d- >n * * * * * * * SO >n in r~- o as in in d- as 00 00 r-H ■d- q oc CO q so CN q q o o q OS CO q q o rH d i-H i—i o o i— H d d CN rH d d d CN CN * * * * * * * * * * * * * * * CN o CN Os r- CN r- d- d- r-H CN Os d- 00 q in q 00 CN i— < q q so q q CN q co 00 d o d d rH d CO CO o rH oo o od so co CO CO CO CO d- 00 * * * * * * r- o SO SO C" o o SO OS CN in co in as q CO CN © t-H i— i CO q oo q CO q CO q os rH o t-H d d d CN CN CN CN SO d d 00 rH t-H * * * * * * * * * * * * * * * * in t-H i— H o o CN d- as O d- d- CN oc CO q q q SO CN o CN oo o q Os q q t-H q d CN H d CO d d d d d 00 rH CO as CN "st- co d" d- CO CN ed cd t-H (N CN r“l CN CN rH CN CN t-H CN CN ,—l 1—1 X X X X X 4) a a> a> CO co 00 cn CO X X . 3 X X £ x .-h o> H- I CO o a O 4-> (X) 4-^ C/3 Sex .S J s h4 13 +-< *H •i a o H u- j. fH 3 £ .5 -J £ CD £ x q > 3 Q g On T— I O'! ^ r- - 3 ^ 2 | 5 6 3 O CO T3 V© ^ ON to ^ -M JS G 4) 0) O CO 3 « 3 IH < a v©vr)vov©vovovomv©v© *o *o *o *o *o *o *0 *o *o o *o o *0 0*0 0 0 0 0*0 0 0 0*0 45 0 o 6 * •a J 1 B to a x «3 <3 co .£ I ^ O Co § a P o ■fc € o 3 0 6 Is 65 & 1 H S ° g CO a a c -a ■*■>* +5 a a *> cq » it CM q in » c o o )-l CL. /— s ■vi- Os vo /— \ /— V O ^ ON /— N c~ <--s in ro / S CM /— s /^■S S>* 00 00 s CN VO / N OO 3 /-N 00 /— s m vo (N CO CM ro t-h q t-H q q CM H t-H q r-H t-H q q r-H q 9 9 9 odd d d d 9 d 9 O 9 9 d 1 d d d , d d t-H 00 Ov OS VO -rfr t-H T—l Os c- OO CM vo vo t-H VO 4 VO 00 in 4 4 m CM CM q CM q q t-h q q t—H t-H q t-H q t-H o q q t-H q d d d 00 o odd CO Ch d d 00 d d VO o d d d c^ cn d d d d d d d d d d S-H W- T—H r- S—T 'w' 'w' ro in o rH 00 Tt CM o Ht 00 VO t-H vo t-H § o o 00 o in o o o oo 00 f- 00 fN t-H On On VO 00 "It CM (N i t-H Ov 4 o ro SO SO c~ r- t-H VO "t c- o in VO ■vt- m fN q O q q t-h p q t-H q q o q t-H t-H t-H q q t-H q t-H o t-H o q q d d d t-H odd t-H d o c4 d d 3 d d d i-H d d d d d d d d o o /— s N S-. /— s /— N /”“■> . /-H /— s /-V c- ■Vf vo •vt 00 Os It CM OV o 00 00 t-H c- iO o 00 VO q ro CM ro t-h ro t-h q ■vt ro t-H t-H q q t-H r—H q 9 q d d d 9 d d d 9 9 9 9 d d O d d CM t-H t-H d 00 m vo CO T-h VO V© ro ro in 4 fN VO ro 9 ro ro q q t-H ro t-H q ro q t-H q CO t-H q o q d d d SD d d d o d d OO 00 o d CM d d d CN d d d d d t-H -■ s*»^' ' CO c- 00 t-H r- in r~ CM m CM n oo in ^ ^ t-h o q q -h q d d d d d d 00 oo o ro in o o Os o 00 00 o t-H ro t-H CM t-H t-H t-H CM o ro CM t-H t-H t-H d d d d d d d d d d d d o CM *n 4 d ro 00 4 00 ro d »n 00 oo ro t-H t-H H* q t-H o q q CM CM q o q CM d d d N d d o o d d SO o o ro d d d yX 00 as oo 1 N" vo o CM r- . CM 00 q ro 00 CM CM i vo r~~ T— I 't- Os 00 00 00 P VO ro 00 vo 00 Os in ro t-H t-H O) t-h o ro q q O q CM as t-H q O q d d d d d d o 4 d d fN d d d d d d 4 vovoovooovoooin oot— looqqo dddddddd ^4oo444voTt oooooqqq dddddddd s_^ ininONinvoinvo-'^- oooooooo dddddddd cm in T-H O oo co O O o vo o co tH O §o & 3 g> £ «S o ^ 4h « ° •§ TO g g O o H h-I C s d ^ g?5 3 £ «S c3 Is Is H H m Co vo v, B ■> o3 B * £ ® Is ^ Ei 3 |J 4 a ^ ^ a> U t a> 2 13 5 So § og S I H £ 4 U CD Jf H- P-. d tE5 £?3 3 ^ 3 3 e £ a> a> Ph Ph d tC c 3 3 £ 2 2 3 § H H c 3 3 ^ d t^ S ^ O > c3 b tfl M B B ^ ^ « « c3 3 S S H H 266 THE JOURNAL OF ARACHNOLOGY Abdomen Figs. 5-8 Protonymph: Four setae on each tergite. First visible sternite (Fig. 5) represents ster- nite 2, setae absent; sternite 3 with two setae; sternites 4 andl 1 each with four setae; all remaining sternites each with six setae; sternites 3 and 4 with one microseta on each stigmatic plate; anal plate with four setae. Pleural membrane granular. Deutonymph: Anterior tergites with fewer setae than posterior ones; some variation in number of setae on each tergite. First visible sternite (Fig. 6) as in protonymph; sternites 3, 4, 5, and 10 with six to 10 setae in a row; sternites 6 to 9 each with six to eight setae on posterior border and two setae set anteriorly on each side of midventral line; one specimen with 10 setae on sternite 9, anterior setae lacking; sternite 11 with four setae; Figs. 5-8. -M. confusum. Sternites 2-7; 5, Protonymph; 6, deutonymph; 7, female; 8, male. (Scale: 1 division = 0.1 mm) NELSON-MORPHOLOGY AND LIFE HISTORY OF MICRO BI SI UM 267 sternites 3 and 4 each usually with one microseta on each stigmatic plate; anal plate as in protonymph. Pleural membrane as in protonymph. Female: Setae of tergites as in deutonymph. First visible sternite (Fig. 7) represents sternite 2, two setae present; genital complex simple, opening at posterior margin of sternite 2; sternites 3, 4, 5, 9, and 10 with eight to 12 setae in a row; sternites 6, 7, and 8 with eight to 13 setae on posterior border and two setae set anteriorly on each side of midventral line; sternite 1 1 with six setae; sternites 3 and 4 each with two microsetae on each stigmatic plate; anal plate as in protonymph and deutonymph. Pleural membrane as in protonymph and deutonymph. Male: Setae of tergites as in deutonymph and female. Genital complex (Fig. 8) with many microsetae; first visible sternite represents sternite 2, 10 microsetae present; an- terior portion of sternite 3 with seven to nine microsetae in a horseshoe shape with two or three microsetae flanking base of horseshoe; posterior margin with seven or eight microsetae in a row. Sternites 4, 5, 9, and 10 with eight to 10 setae in a row along posterior border; sternite 6 with eight to 11 setae on posterior border and two setae set anteriorly on each side of midventral line; sternites 7 and 8 have eight setae each on posterior border with two setae set anteriorly as in sternite 6; sternite 11 with six setae; stigmatic setae as in female; anal plate as in other instars. Pleural membrane as in other instars. Chelicerae Fig. 9-16 Protonymph: Chelicerae (Fig. 9) 1.45-2.12 times longer than wide; cheliceral galea in form of sclerotic knob; cheliceral surface smooth. Movable finger without setae; hand with four setae, es, is, db, and b. Ten to 14 teeth present on each finger; flagellum (Fig. 13) with three or four blades, distal one usually separated from others by width of one blade. Deutonymph: Chelicerae (Fig. 10) 1.4-1.98 times longer than wide; cheliceral galea and surface as in protonymph. Movable finger with one seta,gs; hand with five setae, es, Figs. 9-12.-47. confusum. Chelicera (serrulae omitted): 9, Protonymph; 10, deutonymph; 11, female; 12, male. Abbreviations for setae of the chelicera; galeal seta; es, exterior seta; b, basal seta; s,, sub-basal seta; db, dorsal basal seta; is, interor seta. (Scale: 1 division = 0.1 mm) 268 THE JOURNAL OF ARACHNOLOGY is, db, sb, and b (setae gs and sb not present in protonymph). Ten to 14 teeth present on fixed finger and 12 to 16 teeth on movable finger; flagellum (Fig. 14) with four blades, distal one separated from others as in protonymph. Female: Chelicerae (Fig. 11) 1.48-2.3 times longer than wide; galea, surface and setae as in deutonymph. Thirteen to 18 teeth present on each finger; flagellum (Fig. 15) with four or five blades, distal one separated from others as in protonymph and deutonymph. Male: Chelicerae (Fig. 12) 1.16-1.37 times longer than wide; galea, surface and setae as in deutonymph and female. Thirteen to 15 teeth on fixed finger and 10 to 15 teeth on movable finger; flagellum (Fig. 16) with five or six blades, distal one separated from others as in other instars. Pedipalps Figs. 17-28 Protonymph: Movable finger of chela (Fig. 17) 0.97-1.14 times longer than hand; hand longer than wide; chelal surface smooth; one trichobothrium, t present on movable chelal finger and three on fixed finger, et and eb externally and ist internally. Nineteen to 25 teeth on movable chelal finger and 17 to 22 teeth on fixed finger. Palpal tibia 1.38-2.0 times and palpal femur (Fig. 18) 2.0-2.66 times longer than wide; surfaces of tibia and femur smooth; coxae of pedipalps (Fig. 25) with five setae, two on manducatory process and three on maxilla. Deutonymph: Movable finger of chela (Fig. 19) 0.91-1.25 times longer than hand; hand longer than wide; chelal surface as in protonymph; two trichobothria, t and b present on movable chelal finger and six on fixed finger, et, est, and eb externally and it, ist, and ib internally. Thirty or 31 teeth on movable chelal finger and 28 or 29 teeth on i i Figs. 13-16.-M confumm. Flagellum of the chelicera (stylized). Teeth not always seen in all specimens: 13, Protonymph; 14, deutonymph; 15, female; 16, male. (Scale: 1 division = 0.1 mm) NELSON-MORPHOLOGY AND LIFE HISTORY OF MICRO BISIUM 269 fixed finger. Palpal tibia 1.5-1.81 times and palpal femur (Fig. 20) 2.49-2.8 times longer than wide; surfaces of tibia and femur as in protonymph. Coxae of pedipalps (Fig. 26) with eight setae, three on manducatory process and five on maxilla. Female: Movable finger of chela (Fig. 21) 1.0-1.16 times longer than hand; hand longer than wide; chelal surface as in protonymph and deutonymph; three trichobothria, t, st, and b, present on movable chelal finger and seven on fixed finger, et, est, esb, and eb externally and it, ist, and ib internally; trichobothrium sb on movable chelal finger and isb on fixed finger absent. Thirty-four to 37 teeth on movable chelal finger and 34 to 36 teeth on fixed finger. Palpal tibia 1.61-2.0 times and palpal femur (Fig. 22) 2.57-2.88 times longer than wide; surfaces of tibia and femur as in protonymph and deutonymph. Coxae of pedipalps (Fig. 27) with eight to 10 setae, three on manducatory process and five to seven on maxilla. Male: Movable finger of chela (Fig. 23) 1.01-1.08 times longer than hand; hand longer than wide; chelal surface and number and position of trichobothria as in female. Thirty- two to 36 teeth on movable chelal finger and 30 to 34 teeth on fixed finger. Palpal tibia 1.68-1.93 times and palpal femur (Fig. 24) 2.4-2.78 times longer than wide; surfaces of tibia and femur as in other instars. Coxae of pedipalps (Fig. 28) with nine to 10 setae, three on manducatory process and six or seven on maxilla. Legs Figs. 25-28 Protonymph: Leg coxae I-IV (Fig. 25) each with one seta anteriorly; surface of legs smooth. Deutonymph: Leg coxae I-IV (Fig. 26) each with two or three setae both anteriorly and posteriorly; coxal surface smooth; surface of legs smooth to weakly sculptured. Figs. 17-24.- M. confusum. Chela: 17, Protonymph; 19, deutonymph; 21, female; 23, male. Femur and tibia of the pedipalps (setae omitted). 18, Protonymph, 20, deutonymph; 22, female; 24, male. Abbreviations for the trichobothria. Movable finger: t, terminal; st, subterminal; b, basal. Fixed finger: et, exterior terminal; est, exterior subterminal; esb, exterior sub-basal; eb, exterior basal; it, interior terminal; ist, interior subterminal; ib, interior basal. (Scale: 1 division = 0.1 mm) 270 THE JOURNAL OF ARACHNOLOGY Female: Leg coxae I-IV (Fig. 27) each with two to four setae both anteriorly and posteriorly; coxal and leg surfaces as in deutonymph. Male: Leg coxae I-IV (Fig. 28) each with two or three setae both anteriorly and posteriorly; coxal and leg surfaces as in deutonymph and female. Figs. 25-28. -M. confusum. Coxal area: 25, Protonymph; 26, deutonymph; 27, female; 28, male. (Scale: 1 division = 0.1 mm) NELSON-MORPHOLOGY AND LIFE HISTORY OF MICROBISIUM 111 TAXONOMIC CONSIDERATIONS Microbisium confusum belongs to the family Neobisiidae, subfamily Neobisiinae of the Suborder Diplosphyronida. It can be separated from other Neobisiinae by the presence of only three trichobothria present on the movable chelal finger and seven on the fixed finger. Other Neobisiinae have four trichobothria on the movable chelal finger and eight on the fixed finger. Microbisium confusum can be separated from M. brunneum, in general a much larger species, using the method described by Hoff (1949) and modified by Nelson (1975). With this method M. confusum was considered to have a palpal femur less than 0.42 mm long and a length x width ratio of 2.4-2.93. However, if the length is greater than 0.4 mm, then the length x width ratio is less than 2.8. Microbisium confusum and M. brunneum are sympatric for only part of their ranges. Microbisium confusum is often separated from M. parvulum (Banks), a southwestern U.S. species, by shorter palpal podomeres. However, according to Hoff and Bolsterli (1956) separation is difficult due to overlapping ranges in absolute sizes unless a series of specimens is available. They further, indicate that differences in the shape of palpal podomeres may often work in separation of the species as M . parvulum usually has a less convex extensor margin of the palpal tibia. DISPOSITION OF THE TRICHOBOTHRIA IN RELATION TO GROWTH Gabbutt and Vachon (1968) state that “the time at which each of the trichobothria first appear during development can be summarized by using the method first employed by Vachon (1936). The interval between the oblique strokes represents a stage; proto-, deuto-, tritonymph and adult for the trichobothria of the movable finger (dm) and the internal (dfi) and external (dfe) series on the fixed finger of the chela.” Thus the tricho- bothrial formula for Microbisium confusum and other species of Microbisium is: dm / t / b / st / dfi / ist / it, ib / — / dfe / et, eb / est / esb / The sexually mature individuals bear the arrangement of trichobothria usually present on tritonymphs of other pseudoscorpions species sb and isb, trichobothria usually present in adults of other species, including other Neobisiinae are absent. Gabbut (1965) found the trichobothrial formula in three species of Neobisium to be: dm It / b I st / sb / dfi / ist / it, ib I — / isb / dfe / et, eb / est / esb / — / Therefore the trichobothria of Microbisium and the reported Neobisium appear at the same stage during post-embryonic development through the tritonymph stage; however, the adults of Neobisium possess setae sb and isb . The status of a tritonymph in the genus Microbisium remains unclear. Individuals designated as females and males may be neotenic tritonymphs as indicated by Weygoldt (1969), adults, or a tritonymph-adult complex. A tritonymph-adult complex would con- sist of both tritonymphs and adults grouped in such a way that it would be difficult to distinguish one from the other except by rearing the species in the laboratory. This speculation assumes the adult instar lacks setae sb and isb. The solution to this problem is a course for future study. 272 THE JOURNAL OF ARACHNOLOGY Table 3.-The life stages and number per square meter of Microbisium confusum in Gray Woods during the period 26 March 1976 to 23 November 1977. Date Protonymphs Duetonymphs Females Total 1976 26 March 25 0 51 76 7 April 42 8 25 75 23 April 42 0 25 67 10 May 59 8 34 101 21 May 118 8 59 185 4 June 84 0 42 126 1 8 J une 152 8 34 194 2 July 8 8 42 58 16 July 0 51 8 59 2 August 34 34 8 76 13 August 126 34 34 194 30 August 135 0 84 219 10 September 84 0 42 126 24 September 135 8 93 236 8 October 211 8 101 320 21 October 194 0 8 202 5 November 109 0 34 143 30 November 34 0 25 59 10 December 59 0 59 118 23 December 76 8 143 227 1977 6 January 59 0 59 118 26 January 34 0 8 42 9 February 51 0 34 85 23 February 8 0 8 16 9 March 76 0 34 110 30 March 51 8 8 67 1 3 April 93 0 93 186 27 April 160 17 76 253 12 May 67 0 59 126 25 May 76 0 76 152 9 June 84 0 101 185 22 June 8 8 109 125 6 July 0 8 17 25 20 July 0 76 8 84 3 August 0 17 8 25 17 August 67 8 17 92 31 August 135 0 177 312 14 September 152 17 59 228 29 September 168 8 17 193 14 October 84 8 42 134 26 October 236 8 76 320 9 November 67 8- 8 83 23 November 8 0 42 50 NELSON-MORPHOLOGY AND LIFE HISTORY 0¥ MICROBISIUM 273 LIFE HISTORY A total of 699 individuals (406 protonymphs, 45 deutonymphs, and 248 females) was collected during this study in Gray Woods. No males were found. The population struc- ture of Microbisium confusum in Gray Woods is shown in Table 3. The data are projected into numbers per square meter. Protonymphs were present in all but four collections and reached a peak of 211 per square meter on 8 October 1976 and 236 per square meter on 26 October 1977. Proto- nymphs were absent during at least part of July each year. The protonymphs also showed a spring pulse. Deutonymphs were less abundant and collected on only 23 of 43 dates during this study. The deutonymphs reached a peak of 51 per square meter on 16 July 1976 and 76 per square meter on 20 July 1977. The deutonymphs did not demonstrate the bimodal pulse as seen in the protonymphs. Females were present on all sampling dates. The females reached a peak of 143 per square meter on 23 December 1976 and 177 per square meter on 31 August 1977. Nelson (1973) found that Microbisium confusum in Michigan reached a peak of 155 per square meter and dropped below 20, except for winter months, on a single occasion. The period December through March was interpreted as a “suspended” period with a marked absence of individuals. The absence of individuals was probably due to migration into the soil, to hiberation, or to both. The soil itself was generally frozen beyond the sampling depth. Such an absence of individuals did not occur as markedly in Gray Woods, and in fact the 227 individuals per square meter were found on 23 December 1976 under more than 0.6 meter of snow. Protonymphs and females were collected on all sampling dates throughout the winter months. Deutonymphs probably overwintered but were collected less frequently. Usually the ground was not frozen during the winter months, due to snow cover. However, individuals could still have migrated deeper into the soil and thus not be collected, and as a result the Gray Woods data would compare more favorably with the Michigan data during the winter period. The marked decrease in protonymphs and females during July and August does not appear to be related to a comparable summer “suspended” period due to aestivation, as the deutonymphs reached their peak during this period each year. However, some deutonymphs may have aestivated. This would possibly explain the reason why so few deutonymphs were collected. Nelson (1973) reported a total of 182 individuals during the Michigan study. Of these there were 72 protonymphs, 43 deutonymphs, and 67 females. When compared to Gray Woods the data are summarized in percentages as follows: It is noted that the deutonymphs in Gray Woods represent a marked difference in per- centage. The difference in the percentage of deutonymphs is reflected in the proto- nymphs almost exclusively, as the percentage of females remains similar. The pronounced spring and fall peaks for protonymphs and a somewhat similar situa- tion for the females might indicate more than one generation per year. Nelson (1973) concluded that the data concerning Microbisium confusum from Michigan did not indi- cate more than one generation produced per year. Of six species examined by Gabbutt (1969) only one, Neobisium muscorum , produced more than one generation per year, and this did not hold for all populations of this species. Gabbutt (1970) explained the absence of females during certain periods of the year as being due to their construction of silken chambers for brood purposes. No females with eggs or embryonic stages attached were collected during this study. The cyclic activity of the protonymphs could also be 274 THE JOURNAL OF ARACHNOLOGY explained by the timing as to when they emerged from brood chambers compared to the actual sampling date. A single peak per year for deutonymphs supports a single generation per year for this species. These data do not indicate that more than one generation is produced during a single year. However, females might reproduce in successive seasons. ACKNOWLEDGMENTS I wish to thank Mr. Eben J. Poland of Oswego, New York, for permission to sample in Gray Woods. I am grateful to Mr. Donald Malone for assistance in collecting specimens, to Mr. David Mallow for help in sorting and preparing specimens, and to Mr. David Leece for assistance with illustrations. Special thanks go to Dr. William B. Muchmore who loaned specimens, offered suggestions, and critically reviewed the first draft. For assistance in typing and aid in proofreading the manuscript, I thank Ms. Martha Cavanaugh and Ms. Marian Coffin. LITERATURE CITED Chamberlin, J. C. 1930. A synoptic classification of the false scorpions or chela spinners, with a report on the cosmopolitan collection of the same. -Part II. The Diplosphyronida. Ann. Mag. Nat. Hist., (10) 5: l-48;585-620. Gabbutt, P. D. 1965. The external morphology of two pseudoscorpions Neobisium carpenteri and Neobisium maritimum. Proc. Zool. Soc. Lond., 145:359-386. Gabbutt, P. D. 1969. Life histories of some British pseudoscorpions inhabiting leaf litter. In The Soils Ecosystem, ed. J. G. Sheals. Systematic Assoc. Publ., 8:229-235. Gabbutt, P. D. 1970. Sampling problems and the validity of life history analyses of pseudoscorpions. J. Nat. Hist., 4: 1-15. Gabbutt, P. D. and M. Vachon. 1968. The external morphology and life history of the pseudoscorpion Microcreagris cambridgei. J. Zool. Lond, 154:421-441. Hagen, H. 1869. The American Pseudo-scorpions. Record of American Entomology for the Year 1868:48-52. Salem. Hoff, C. 1946. American species of the pseudo scorpion genus Microbisium Chamberlin, 1930. Bull. Chicago Acad. Sci., 7:493-497. Hoff, C. 1949. The pseudoscorpions of Illinois. Bull. 111. Nat. Hist. Survey, 24:407-498. Hoff, C. and J. E. Bolsterli. 1956. Pseudo scorpions of the Mississippi River drainage area. Trans. Am. Microsc. Soc., 75:155-179. Lawson, J. E. 1969. Description of a male belonging to the genus Microbisium (Arachnida: Pseudo- scorpionida). Bull. Virginia Polytechnic Institute. Research Div., 35:1-7. Nelson, S., Jr. 1973. Population structure of Microbisium confusum Hoff in a beech-maple woodlot. Revue d’Ecologie et de Biologie du Sol., 10:231-236. Nelson, S., Jr. 1975. A systematic study of Michigan Pseudoscorpionida (Arachnida). Am. Midi. Nat., 93:257-301. Vachon, M. 1936. Sur le developpment post-embryonnaire des Pseudoscorpiones (4 erne note). Less formules chaetotaxiques des pattes-machoire. Bull. Mus. Hist. Nat., Paris, (2) 8:77-83. Weygoldt, P. 1969. The biology of pseudoscorpions. Harvard University Press, Cambridge XIV +145 pp. Manuscript received May 1981, revised December 1981. The Journal of Arachnology 10:275 RESEARCH NOTES EURYOPIS COKI (THERIDIIDAE), A SPIDER THAT PREYS ON POGONOMYRMEX ANTS During studies of western harvester ant populations (Porter and Jorgensen 1980), we frequently observed the spider Euryopis coki Levi closely associated with colonies of the ant, Pogonomyrmex owyheei Cole. These observations indicated that E. coki was a specialized predator of P. owyheei workers. In his taxonomic revision of the genus Euryopis , Levi (1954) listed several instances of these spiders preying on ants. Gertsch (1979) stated that Euryopis spiders “are reputed to feed largely on ants but undoubtedly also prey on other small insects. On one occasion”, he watched a female of Euryopis texana Banks “prey upon a moving line of small ants, grasping and dispatching them in numbers until the rapacious spider had gathered a small heap of victims.” Berland (1933) reported that Euryopis acuminata (H. Lucas) of Europe captured Crematogaster ants and transported them attached to its spinnerets. Carico (1978) described in detail the feeding habits of Euryopis funebris (Hentz) in Virginia. He noted that E. funebris first tethered Camponotus ants with loops of viscid silk and then killed them with a bite on the leg. The purpose of our paper is to provide an account of the interaction between E. coki and the harvester ant ,P. owyheei. E. coki ranges across southern Idaho into northern Utah and western Wyoming (Levi 1954), perhaps also extending into Oregon, Nevada and Colorado. We observed E. coki in the Raft River Valley of southern Idaho during June, July and August, 1977-1980. These spiders were commonly found associated with P. owyheei in sagebrush ( Artemisia tri- dentata ), greasewood (Sarcobatus vermiculatus) and shadscale (A triplex con ferti folia) plant communities. The frequency of spider observations generally increased as the sum- mer progressed so that by early August, two to three spiders were often present on each ant mound. Occasionally, as many as a dozen spiders were observed at a single mound. Spiders were usually most numerous on mounds with little ant activity, suggesting either that large numbers of foraging ants prevent the spiders from gathering on the mounds or that the spiders may actually cause the mound’s reduced activity in much the same way as predation behavior reported for the horned lizard Phrynosoma cornu turn (Whitford and Bryant 1979). Individual spiders were most often found hiding among small rocks on the surface of the ant mound where their size and cryptic color pattern made them difficult to see. In the evenings and on other occasions when the ants had temporarily plugged their mound entrance with small stones, E. coki was often found lying in wait directly over the shallow depression that remained. Occasionally when an ant mound was constructed near a dead bush, the spider retreated to the lower branches where the bodies of its prey would be left dangling on short silken threads (2-5 mm). Perhaps E. coki occasionally employs a “dangling feeding” behavior similar to that reported for E. funebris (Carico 1978). More The Journal of Arachnology 10:276 commonly, dead ants were simply found abandoned around the mound periphery. Prey usually consisted of harvester ants, but on occasion small grasshoppers and other species of ants were found dangling from branches as if they had been taken by the spider. Spiderlings of this species were not observed on the ant mounds, possibly because they smaller arthropod prey, perhaps smaller species of ants as has been reported for E. funebris (Carico 1978). The following account of E. cokf s feeding behavior was recorded in early August 1977. “Upon returning a recapture sample of ants to their mound late in the afternoon, several small gray spiders were observed which had been seen earlier in the morning. One of the largest individuals attacked a passing worker by first tacking it to the ground with a short strand of silk and then biting it on the leg. Initially, the ant swung back and forth as if it had been tethered on a leash, but shortly it lapsed into convulsions and ceased motion within two minutes. This disturbance attracted a second and a third worker in succession; both briefly attempted to attack the spider before they were temporarily entangled by silk and driven away. Meanwhile, eight to nine additional spiders of both sexes were observed alternately running on the mound and then lying in wait among the rocks. Over the next 20 minutes nearly every spider managed to kill an ant. After death, the ants were usually carried several decimeters off the mound in a peculiar sling fashion tied to the tip of the spider’s abdomen” (Fig. 1). This behavior allowed the spiders to consume their prey without further disturbance from other ants. Predatory behavior of E. coki is quite similar to that observed for E. funebris (Carico 1978) except that E. funebris is primarily nocturnal and arboreal while E. coki is primar- ily diurnal and ground dwelling. The spiders Steatoda fulva (Keyserling) and Zodarium frenatum (Simon) are also similar to E. coki in that they capture ants from ant nests. S. fulva captures Pogonomyrmex badius (Latreille) workers in webs which are constructed over the mound entrances during periods of worker inactivity (Holldobler 1970). Z. Fig. l.-Euryopis coki female transporting a worker of the harvester ant Pogonomyrmex owyheei (drawing by Jean A. Stanger). The Journal of Arachnology 10:277 frenatum captures Cataglyphis bicolor (F.) workers at night by lying in wait near the nest entrance and attacking the guard ants (Harkness 1977). In early August 1980, we collected approximately 50 spiders for laboratory observa- tion of feeding behavior. The mean dry weight of 40 ants fed to these spiders was 2.1 ± 0.4 mg (±SD), while the mean dry weight of 60 ants not fed to the spiders was 2.4 ± 0.3 mg. Apparently, the spiders removed about 13% (0.3 mg) of the ants’ biomass during feeding. The mean live weight of 12 spiders before feeding was 3.5 ± 1.5 mg. After feeding, the spiders’ weight had increased an average of 1.1 ± 0.9 mg or 30 %. Twelve egg sacs were deposited during the course of these observations with multiple layings by several females. Eight egg sacs were allowed to hatch and they averaged 19.8 ± 4.7 spiderlings. The spiderlings emerged about 19 days after the egg sacs were deposited. E. coki appears to be a fairly important predator of P. owyheei in southern Idaho, especially in late July and early August. While our observations suggest that these spiders specialize on harvester ants, it is possible that other major food sources exist. Since several closely related Euryopis species inhabit southwestern United States (Levi 1954), it would be interesting to know what relationship these spiders have with ants in their regions. Thanks are extended to J. E. Carico and C. D. Jorgensen for their helpful suggestions. W. J. Gertsch and H. W. Levi identified the spiders and read the manuscript. Voucher specimens have been deposited in the Museum of Comparative Zoology (MCZ) at Harvard University. This project (Report no. DOE/ID/01 674-7) was sponsored by EG&G Idaho, Inc. and partially funded by DOE contract no DE-AS07-77ID01674 to Brigham Young University. Additional funding was provided by Brigham Young University Department of Zoology and the D Eldon Beck Award. LITERATURE CITED Berland, L. 1933. Contribution a fetude de labiologie des arachnides. Arch. Zool. Exper., 76(1): 1-23. Carico, J. E. 1978. Predatory behavior in Euryopis funebris (Hentz) (Araneae: Theridiidae) and the evolutionary significance of web reduction. Symp. Zool. Soc. Lond., 42:51-58. Gertsch, W. J. 1979. American Spiders. 2nd ed. Van Nostrand Reinhold, New York, 274 pp. Harkness, R. D. 1977. Further observations on the relation between an ant Cataglyphis bicolor (F.) (Hym., Formicidae) and a spider, Zodarium frenatum (Simon) (Araneae, Zodariidae). Entomol. Mon. Mag., 112:111-121. Holldobler, B. 1970. Steatoda fulva (Theridiidae), a spider that feeds on harvester ants. Psyche, 77:202-208. Levi, H. W. 1954. Spiders of the genus Euryopis from North and Central America (Araneae, Ther- idiidae). Amer. Mus. Novitates, No. 1666, 48 pp. Porter, S. D. and C. D. Jorgensen. 1980. Recapture studies of the harvester ant, Pogonomyrmex owyheei Cole, using a fluorescent marking technique. Ecol. Entomol., 5:263-269. Whitford, W. G. and M. Bryant. 1979. Behavior of a predator and its prey: The horned lizard ( Phrynosoma cornutum ) and harvester ants ( Pogonomyrmex spp.). Ecology, 60:686-694. Sanford D. Porter and David A. Eastmond, Department of Zoology, Brigham Young University, Provo, Utah 84602 (Present address of senior author: Department of Bio- logical Science, Florida State University, Tallahassee, Florida 32306). Manuscript received March 1981, revised July 1981 The Journal of Arachnology 10:278 NEW SENSORY (?) ORGAN ON A SPIDER TARSUS While examining Australian spiders of the general Arcys and Arche moms (family Araneidae), S. H. discovered a new structure. On the dorsal surface of tarsus I the males exhibit a brush-like structure composed of about 1500 fine and very short hairs. In all the species examined, these hairs almost totally cover the upper side of the tarsus. Scanning electronic microscope examination revealed setae of an unusual shape (Fig. 1). It is likely that the hairs are chemosensory structures, although neither the paper of Foelix and Chu-Wang (1973. Tissue and Cell 5(3):46 1 -47 8) nor of Kronestedt (1979. Zool. Scripta, 8(4): 279-285) mention similarly shaped sense organs. It may be assumed that the new structure is of importance in courting, or in copulation, since the brush is found only in mature males— females and last instar juvenile males have normal setae at that location. At present, research has not revealed any corresponding structures on the females, such as Fig. 1.— a. first tarsus of male Archemorus roosdorpi Chrysanthus; scale line 500 jam long. b-c. Setae from brush-like structure on tarsus, scale lines 10 jum long. d-e. Distal ends of individual setae, scale lines 1 Atm long. The Journal of Arachnology 10:279 fields of glands which might interact with the male sense organs. Studies of courtship behavior and perhaps a histological examination with fresh material, are needed to clarify the nature and function of this interesting structure. We would like to thank Dr. R. F. Foelix for advice on the possible function; Ed Seling for help with the scanning electron microscope; and the Freshman Seminar Program of Harvard University and the National Science Foundation grant DEB 80-20492 to H. W. L. for financial assistance . Stefan Heimer, Staatlichen Museum fur Tierkunde, Dresden, Deutsche Demokratische Republik; John M. Hunter, Museum of Comparatice Zoology and Timothy S. Oey, Har- vard College; and Herbert W. Levi, Museum of Comparative Zoology, Cambridge, Mass. 02138. Manuscript received January 1982, revised February 1982. ADDITIONAL OCULAR ANOMALIES IN SPIDERS I first published a short note on ocular anomalies in 1937. Since the publication of my 1962 paper a few other accounts dealing with ocular anomalies have appeared, and six additional specimens have come to my attention. I had referred to cases where spiders of hypogean habits had had their eye size reduced, or where one or more eyes were com- pletely lost. Sanocka recently (1981) reported the results of her study with one species of this kind, Porrhomma moravicum Miller and Kratochvil. She discussed and illustrated the 12 different kinds of anomalous situations she encountered. As to teratological specimens, later in the same year as my own paper Boggild (1962) reported for a specimen of Asagena phalerata (Panzer), that it was “lacking the posterior median eye of the right side, no trace of it being visible on the surface.” Engelhardt (1964) reported for Trochosa ruricola (DeGeer) one female with both anterior lateral eyes and the right posterior lateral eye missing, and still another female completely eyeless. In 1968 I called attention to a young spiderling of Latrodectus hesperus Cham- berlin and Ivie with 16, and another with 14 eyes. Both of these cases were quite obviously the results of embryonic duplication of a portion of the head region. Marer (1972) described the situation in a tarantula, Aphonopelma reversum Chamberlin, in which only the left posterior lateral eye appeared normal. In 1972 I suggested that the four-eyed spider described by Rafinesque (1821) as a new species was in all probability an anomalous lycosid, and I pointed out that the literature contains more references to eye anomalies in lycosids than in any other spiders. I am adding herewith still another seen by me, and also one of which I have learned. Case No. 1.— Juvenile lycosid: This specimen was collected by A. Loveridge at Lumbo, Mozambique on 18 July 19 . The left anterior lateral eye is missing but there is a little dark pigment at the locus of the missing eye. In addition, there is somewhat less pigment between the posterior median and lateral eyes of the left side compared with the right (Fig. 1). The Journal of Arachnology 10:280 Figs. 1-6.- 1, Lycosid from Mozabique, eye area from in front; 2, Sparassid “a”, eye area from above; 3, Sparassid “b”, eye area from above; 4, Rualena cockerelli, eye area from in front, 5, Bathyphantes concolor, eye area from above; 6, Araneus pratensis, eye area from above. Case No. 2.— Sparassid “a”: This specimen, a female, was collected by George Schwab in Cameroon (Date not available). All the eyes appear normal except the anterior me- dians. The right one is shifted to the left of its normal position, and the left one is reduced in size, its diameter being less than half that the the right one (Fig. 2). Case No. 3.— Sparassid “b”: This is another female with the same collection data as for No. 2. In this specimen the four eyes of the right side are normal. On the left side the anterior median eye is reduced in size, being less than half the diameter of the right eye. The left anterior lateral eye is much smaller than its mate on the right side, has a much less distinct lens, and has less surrounding dark pigment. The left posterior lateral eye is completely missing (Fig. 3). Case No. 4 —Rualena cockerelli Chamberlin and Ivie: This is a male collected by V. D. Roth about 18 km south of Tecate, Baja California, 10 November 1957. In this specimen there are only three eyes present, namely, the anterior lateral, the posterior lateral and the posterior median of the right side. At the locus of the left posterior median eye there is some dark pigment both in front of, and behind, where the eye would be if it had developed. In addition, compared with a normal set of eyes one can see that both of the right lateral eyes are reduced in size, each being only about half the diameter of the normal (Fig. 4). Case No. 5.— Bathyphantes concolor (Wider): This is a female collected by J. F. Ander- son at Hartford, Connecticut, 10 March 1961. The left posterior median eye is complete- ly missing and there is no dark pigment at the locus where the eye normally appears. Moreover, there is less pigment around the left anterior median eye than around the corresponding right eye (Fig. 5). Case No. 6 —Araneus pratensis (Emerton): This is a female collected by J. F. Anderson at Newington, Connecticut 14 June 1961. The left posterior lateral eye is completely missing and there is no dark pigment to indicate its locus. Moreover, the amount of pigment around the left anterior lateral eye is much less than for the corresponding right eye (Fig. 6). Another case.— In addition to the above six cases seen by me I can report the following of which I have heard (pers. comm. V. D. Roth). An immature male Lycosa santrita Chamberlin and Ivie collected in Pima County, Arizona, 7 July 1973 by J. and F. Murphy possessed only the right anterior lateral eye, all seven other eyes being lacking. It was The Journal of Arachnology 10:281 noted also that the eye was reduced in size, its diameter being less than half that of a normal anterior lateral eye. I thank V. D. Roth and H. W. Levi for making available some of the specimens here described. LITERATURE CITED Boggild, O. 1962. Spiders from Bommerlund Plantation, a spruce forest in South Jutland. Entom. Meddel., 31:225-235. Engelhardt, W. 1964. Die mitteleuropaischen Arten der Gattung Trochosa C. L. Koch, 1848 (Araneae, Lycosidae). Morphologie, Chemotaxonomie, Biologie, Autoekologie. Z. Morph. Oekol. Tiere, 54:219-392. Kaston, B. J. 1937. Structural anomalies in spiders. Bull. Brooklyn Entom. Soc., 32:104. Kaston, B. J. 1962. Ocular anomalies in spiders. Bull. Brooklyn Entom. Soc., 57:17-21. Kaston, B. J. 1968. Remarks on black widow spiders, with an account of some anomalies. Entom. News, 79:113-124. Kaston, B. J. 1972. On Tessarops maritima, a nomen oblitum in spiders. Entom. News, 83:117-118. Marer, P. J. 1972. An eye defomrity in a tarantula spider, Aphonopelma reversum. Pan-Pacific Entom., 48:221-225. Rafinesque, C. S. 1821. Description d’une araignee qui constitute un genre nouveau. Ann. Gen. Sci. Phys. (Bruxelles), 8:88-89, pi. cxcvi, f. 3. Sanocka, Elzbieta. 1980. Eyes regression in Porrhomma moravicum Miller et Kratochvil 1940. Verh. 8 Intern. Arachnol.-Kongr. Wien 1980, p. 383-387. B. J. Kaston, Department of Zoology, San Diego State University, San Diego, CA 92182 Manuscript received November 1981, revised January 1982. M ANTISPA IN A PEUCETIA EGG CASE During a study of reproductive effort in the green lynx spider, Peucetia viridans, one egg sac that was opened contained a second sac neatly filling the entire space within the first. Upon opening this sac, a grub-like larva was found. This larva was allowed to develop within the sacs. The sub-imago that emerged was a mantispid, Mantispa sp. (possibly interrupta). Since the mantispid did not live to expand its wings, species identi- fication could not be made, see key by Froeschner (1947, Ann. Ent. Soc. Amer. 40:123-236) or by Rehn (1938, Trans. Amer. Ent. Soc. 65:237-266). Mantispids undergo hypermetamorphosis; thus the scarabaeiform larva found in the sac had prepared for pupation by feeding on the lynx spider’s eggs. The second sac was the silken cocoon spun by the larva prior to pupation (Borror, D. J., Delong, D. M., Triplehorn, C. A. 1976. Study of Insects. New York. Holt, Rinehart and Winston). A number of authors have indicated that mantispid larvae feed on the eggs of spiders. Borror, et. al., (Ibid. 1976) indicate they are found “in the egg sacs of ground spiders.” Withycombe (1924, Trans. Royal Ent. Soc. London 72:303-411) mentions they attack The Journal of Arachnology 10:282 the egg sacs of Lysosa , and Gertsch (1979, American Spiders. New York. Van Nostrand Reinhold Co.) includes the agelenids along with the lycosids as a mantispid host. Kaston has also described mantispids from agelenids (1938, J. New York Ent. Soc. 46:147-153) and gnaphosids (1940, Bull. Brooklyn Ent. Soc. 35:21). A clubionid sac containing a mantispid is described by Milliron (1940, Ann. Ent. Soc. Amer. 33:357-360), and salticid sacs that produced two pupae of M. interrupta are reported by Smith (1934, J. Kansas Ent. Soc. 7:120-145). Askew (197 1 , Parasitic Insects. New York. American Elsevier Publ. Co.) lists the following host families of mantispids: Gnaphosidae, Clubionidae, Thomi- sidae, Lycosidae and Pisauridae. The present observation appears to be the first record of a mantispid found in the Oxyopidae. It appears that the phenomenon that was observed here is best described as partial brood parasitism. Brood parasitism occurs when a female lays her eggs in the nest (egg sac) of another species, and the young are raised by foster parents. With mantispids, the eggs are laid on vegetation and the first larval stage seeks the egg sac of a spider where it subsequently feeds on the eggs of the spider. Since the female lynx spider guards her egg sac with the mantispid inside the sac, the mantispid benefits from the care this “foster” parent is giving, i.e., sac maintenance and defense of this webbed territory. This behavior is similar to bird behavior where the brood parasites in a host nest are cared for by the parents at the expense of their own young. The term partial brood parasitism is used here because there are some basic differences between the mantispid behavior and that of brood parasitism in birds. First, the female mantispid does not lay her eggs directly into the egg sac; birds do lay their eggs directly into the host’s nest. Second, the female lynx spider does not feed the mantispid larvae as a bird host feeds the young of its social parasite. There are also strong similarities to brood parasitism. The fact that many bird brood parasites kill the host’s young parallels the death of the eggs or spiderlings caused by the feeding behavior of the mantispid larva. Although the female lynx spider does not feed the mantispid larva directly, she does provide food indirectly in the form of her own reproductive output, her clutch. Since the phenomenon observed here includes features found in the basic definition of brood parasitism, i.e., behavioral care by a foster parent, elimination of the hosts’ young, and provision of food by the foster parent, the idea of partial brood parasitism appears to be appropriate. Don W. Killebrew, Department of Biology, The University of Texas at Tyler, Tyler, Texas 75701 . Manuscript received April 1981, revised September 1 981. The Journal of Arachnology 10:283 THE JOURNAL OF ARACHNOLOGY Instructions to Authors GENERAL COMMENTS Manuscripts are acceptable in English, French, Portuguese, and Spanish, and must be typed double or triple spaced throughout. Use only paper 8.5 by 11 in. in size, but not erasable bond. Leave margins at least 1.5 in. on the left and 1 in. on the top, bottom, and right, and do not hyphenate any words at the right margin. To facilitate prompt review by two or more referees, send the Associate Editor the original manuscript and two good copies, together with copies of the illustrations, for mailing to referees. Follow the recommendations of Council of Biological Editors (CBE) Style Manual , Third Edition, unless indicated otherwise below Do not edit your own manuscript: iltalics are permitted only to indicate scientific names; only the TITLE, PRIMARY HEADINGS (e. g. INTRODUCTION, etc.), and RUNNING HEAD should be typed in capital letters. Manuscripts longer than about 1500 words (five double-spaced typewritten pages in Elite, six double-spaced typewritten pages in Pica) should be prepared as Feature Articles, shorter papers as Research Notes. Manuscripts and all related correspondence must be sent to Dr. B. J. Kaston, Associate Editor, 5484 Hewlett Drive, San Diego, California 92115, U.S.A. FEATURE ARTICLES Arrange the various parts of feature articles in the following sequence: (1) mailing address, (2) title, (3) by-line, (4) body of text, (5) acknowledgments, (6) literature cited, (7) figure legends, (8) abstract, (9) footnotes, (10) running head, (11) tables with legends, (12) figures. Put only numbers 1, 2 and 3 above on page 1, and start page 2 with number 4 above, numbering all other pages consecutively. Mailing address. -Include the complete address and the telephone number of that author with whom all correspondence with the Editor should be handled. Title. -When using common or scientific names in the title be sure to include in parenthesis the appropriate higher taxa (order, family,. . .) Include footnote indication if appropriate (e. g., to acknowledge grant support), but type footnote separately (see number 9 above). Title should be typed in capital letters and may not exceed 48 characters and spaces per line. By-line. -Include name(s) of author(s) as customarily written (less titles) and complete address(es) including zip code or other postal zone designation. Include footnote indication(s) if appropriate (e. g., to indicate change of address), but type footnote separately (see number 9 above). Leave ample space between title and by-line, and four spaces between name(s) and address(es). Body of text. -Use whatever form seems best to accommodate the necessary description of re- search. Be concise. Conform to CBE Style Manual in all abbreviations. Use the metric system on all measurements (the English system is acceptable only when transcribing locality data accompanying museum specimens), and note that abbreviations of metric units of measurement are not punctuated (e.g., mm and km, but ft. and mi.). Citations should be in the form: Bellrose (1950), or Bellrose (1950:33), or (Bellrose 1950). The complete scientific name of a species or genus, including author(s), must be given the first time they are mentioned in the text. Use single-line notation for fractions (e. g., 1/4 and not L^(4-12)/3 and not 4-12). 4 3 The following special directions apply to authors of taxonomic papers: (a) . Do not use abbreviations to indicate that a new name or a new combination is being proposed in a primary heading (e. g., A-us x-us, new species, rather than A-us x-us, n. sp. or comparable abbrevia- tions). (b) . Keys myst be typed as follows: 1. Use Arabic numerals to designate the leading entry of a couplet 2 Do not designate the second entry of a couplet, either by means of numbers, letters, or other marks 3 2. Type numbers flush to left margin, and start entry on fifth space. The second, and subsequent, lines of one entry must also be indented five spaces species The Journal of Arachnology 10:284 (c) . Synonymies must follow the abbreviated style shown below: A-us x-us Jones 1930:3, 1935:9; Russell 1945:453; Smith 1954a: 16, 1954b:678; Cooper and Lim 1955:18 (in part). A-usy-us Bates 1932:18, fig. 4. NEW SYNONYMY. A-us z-us : Miranda 1948:98 (misidentification); Harris 1951:3 (in part ?). (nec A-us z-us Zimmer (d) . Lists of specimens examined of a given taxon must be the last item typed in the treatment of that taxon as they will be set in smaller type. Adhere to the following style for listing specimens examined: Country: state or comparable political subdivision; county or district, detailed locality (elevation), 14 July 1945 (collector), 2 males, 5 females (acronym of institution where specimens are deposited), next detailed locality within that county, and so forth; next county in the same state; and so forth: next state in the same country: and so forth. Next country: and so forth. Punctuation rules are very simple. Use a period to separate countries, colon to separate states, semi-colon to separate counties, and commas to separate specific localities. Acknowledgments.- Avoid overlooking persons who have in some substantial way assisted with the work. Authors of taxonomic papers should spell out the name, and indicate parenthetically the acronym, of institutions where specimens studied are deposited. Literature cited.- Include only those publications to which reference is made in the text. Adhere to the CBE Style Manual or refer to a previous issue of The Journal of Arachnology for style. Do not abbreviate place names in journal citations. Repeat name(s) of author(s) in case of multiple entries. Figure legends.- Provide one legend for each illustration to be reproduced singly, or for each “plate” consisting of several illustrations. Adhere to the following styles: Figs. 1-4.- A-us x-us, male from Timbuktu: 1, left leg; 2, right chelicera; 3, dorsal aspect of genitalia; 4, ventral aspect of abdomen. Figs. 27-34. -Right chelicerae of species of A-us from Timbuktu: Figs. 27, 29, 31, 33. -Dorsal views; Figs. 28, 30, 32, 34. -Prolateral views of movable finger; Figs. 27-28: A-us x-us, holotype male; Figs. 29-30: A-us w-us male; Figs. 31-32: A-us z-us, holotype male; Figs. 33-34: A-us t-us, male. Scale = 1.0 mm. Type all figure legends consecutively on same page(s), using double space within each legend and leaving 4 spaces between legends. Keep in mind that 85 characters and spaces represent one printed line; for each line subtract 4 mm from the maximum length permissible on full-page illustra- tions. Abstract.- Abstracts should be a summary of the basic findings, and should not exceed 2 to 3 percent of the text in length. Papers in a language other than English must be accompanied by an English abstract as well as an abstract in the language of the text. Footnotes.-Footnotes will be permitted on page 1 when it is appropriate to acknowledge grant support and to indicate change of address, etc. No footnotes will be permitted on any other page. Type footnotes on a page by themselves. Running head.- At the top of each right hand page will be printed the author(s)’ last name(s) and an abbreviated title. This running head may not exceed 55-60 characters and spaces, must be typed in capital letters, and a hyphen must separate the name(s) from the title. Tables with legends.-Prepare all tables precisely as they are to be typeset. Construct tables as simply as possible. Include legend at the top of each. Make marginal notations in the text which clearly indicate the appropriate point of insertion of each table. Note that in these instructions “graphs” are regarded as “figures,” not “tables.” The size of the printed page imposes a limit on the size of tables that can be accepted. On a normal, upright table these limits are 99 characters and spaces per line and 56 lines long, including legend. Tables printed sideways on a page can be up to 158 characters and spaces per line and 36 lines long, including legend. Sideways tables are difficult to typeset and present problems during paste-up; therefore, they will be accepted in exceptional cases only, as most tables that are “too wide” can usually be re-arranged to fit into the more desireable, upright position. Any table that exceeds the dimensions given above will not be accepted for publication. Tables must be typed double spaced throughout, and the legend must be in the following style: Table 1.- Incorporate into legend all information necessary to make the data presented clearly understandable. Do not resort to use of footnotes. The Journal of Arachnology 10:285 Figures.- Figures are the only parts of the paper for which the author is responsible for preparing camera-ready material, and they must be done with great care. Drawings should be done with black ink on high quality, smooth, white, heavy paper. Drawings done in some manner not employing black ink (e. g., shaded pencil drawings) are difficult to print, more so than photographs. When such a technique is used, the drawings should be fairly dark, have good contrast, and be well executed. The largest dimensions on the printed page to which illustrative material (with legends) will be reduced is 127 by 200 mm. Decisions as to size of originals, placement of several smaller drawings on a single page, etc., should be based upon this format. Where several smaller drawings are to appear on a single page, it is the responsibility of the author(s) to furnish an assembled “plate” precisely as it should appear in print. The “plate” must be prepared to allow for inclusion of the legend beneath the drawings. Number all drawings consecutively using Arabic numerals, do not designate illustrations with letters and never designate plates. Labelling should be done neatly, using either pressure sensitive characters or a lettering apparatus employing black ink. Photographs should have good contrast, be glossy or semi-glossy, and printed slightly larger (up to 2X) than anticipated published size. Illustrative material of any kind larger than two times the anticipated printed size will not be accepted. Larger illustrations are difficult to handle and get easily damaged or lost in the mail. Authors preparing larger illustrations should consult local photo shops and printing firms about the availability of PMT’s (PhotoMechanical Transfer) for reducing their illustrations to acceptable size. PMT’s are high contrast, direct positive-to-positive reproductions on semi-glossy paper which will also give the author(s) a good idea of how the illustrations will look when printed (The Journal of Arachnology does not provide proofs of illustrations). PMT’s can be obtained at a reasonable cost in a number of sizes, usually 8 by 10 in. or 9 by 12 in., with any amount of figure reduction desired within that space. Sending PMT’s rather than originals avoids the loss of art work in the mail, and if the PMT’s get lost or damaged they can be easily replaced. Illustrations, PMT’s, and photographs must be mounted on hard white mount- ing board (Scotch Magic Transparent Tape can be used). Illustrative material that is poorly prepared, too large, or not mounted properly will be returned to the author. RESEARCH NOTES Arrange the various parts of your research notes in the following sequence: (1) mailing address, (2) title, (3) body of text, (4) by-line, (5) figure legends, tables with legends, and illustrations. Follow instructions given above for feature articles unless otherwise indicated below. Do not include foot- notes anywhere, references to grant support and all other acknowledgments should be made in a statement in the body of the text. If an indication of change of address is desired, it should be included parenthetically after the credited institution as “present address.” If five or fewer references are cited in the text, these should be made parenthetically where they appear in the text following the style set for feature articles but omitting the title of the article cited (e. g., Johnes, J. 1967. J. Arachnol., 2:199-214). If more than five citations are to be made, however, separate them from the body of the text and include them in their own section as in feature articles. The by-line must be typed in paragraph form after the body of the text, or the literature cited when one is present. . THE AMERICAN ARACHNOLOGICAL SOCIETY President: Jonathan Reiskind (1981-1983) Department of Zoology University of Florida Gainesville, Florida 32601 Membership Secretary : Norman I. Platnick (appointed) American Museum of Natural History Central Park West at 79th Street New York, New York 10024 Secretary: William A. Shear (1980-1982) Department of Biology Hampden-Sydney College Hampden-Sydney, Virginia 23943 President-Elect: Susan E. Riechert (1981-1983) Department of Zoology University of Tennessee Knoxville, Tennessee 37916 Treasurer: Norman V. Horner (1981-1983) Department of Biology Midwestern State University Wichita Falls, Texas 76308 Directors: Herbert W. Levi (1981-1983) Michael E. Robinson (1981-1983) Vincent D. Roth (1980-1982) The American Arachnological Society was founded in August, 1972, to promote the study of the Arachnida, to achieve closer cooperation between amateur and professional arachnologists, and to publish The Journal of Arachnology. Membership in the Society is open to all persons interested in the Arachnida. Annual dues are $20.00 for regular members, $15.00 for student members. Correspondence con- cerning membership in the Society must be addressed to the Membership Secretary. Members of the Society receive a subscription to The Journal of Arachnology. In addi- tion, members receive the bi-annual newsletter of the Society, American Arachnology. American Arachnology , edited by the Secretary, contains arachnological news and comments, requests for specimens and hard-to-find literature, information about arach- nology courses and professional meetings, abstracts of the papers presented at the Society’s meetings, address changes and new listings of subscribers, and many other items intended to keep arachnologists informed about recent events and developments in arach- nology. Contributions for American Arachnology must be sent directly to the Secretary of the Society. The Eastern and Western sections of the Society hold regional meetings annually, and every three years the sections meet jointly at an International meeting. Information about meetings is published in American Arachnology , and details on attending the meetings are mailed by the host(s) of each particular meeting upon request from interested persons. CONTENTS THE JOURNAL OF ARACHNOLOGY VOLUME 10 FALL 1982 NUMBER 3 Feature Articles The genera Ideobisium and Ideoblothrus , with remarks on the family Syarinidae (Pseudoscorpionida), William B. Muchmore 183 The life history of Centruroides gracilis (Scorpiones, Buthidae), Oscar F. Francke and Steven K. Jones 223 Sex pheromones in two orbweaving spiders (Araneae, Araneidae): An experimental field study, Cader W. Olive 241 Coxal relocation following the loss of an adjacent coxa in Latrodectus variolus (Walckenaer), John B. Randall 247 Spiders in eastern Texas cotton fields, D. A. Dean, W. L. Sterling and N. V. Homer 251 The external morphology and life history of the pseudoscorpion Microbisium confusum Hoff, Sigurd Nelson Jr 261 Research Notes Euryopis coki (Theridiidae), a spider that preys on Pogonomyrmex ants, Sanford D. Porter and David A. Eastmond 275 New sensory (?) organ on a spider tarsus, Stefan Heimer, John M. Hunter, Timothy S. Oey and Herbert W. Levi 278 Additional ocular anomalies in spiders, B. J. Kaston 279 Mantispa in a Peucetia egg case, Don W. Killebrew 281 Others Instructions to Authors 283 Cover illustration, Alacran tartarus Francke, by Oscar F. Francke Printed by The Texas Tech University Press, Lubbock, Texas Posted at Crete, Nebraska, in November 1982 *10783 I 1 £ 1 ^ a a n2 li b r ar i es^smithsonian2 iNsmuTiON^NoiiruiiSNi^NviNosHiiws^ss i ava a lo ^ co — _ co ! < (If j|) i r „ _ .. _ \ A k A / anj v o Z o JTION^MOlinillSNI^NVINOSHimS S31BVB8n LIBRARIES SMITHSONIAN INSTITUTIC 2 ___ v g *“ 2 “ QX O m ^ 2 m ° m V.W 2r m x^osv^ ^ m yan LIBRARIES SMITHSONIAN^INSTITUTION^OIlfUllSNI^NVINOSHillNS^ S3IHVU8 z > z \ co z / z JgJF > '^E«2^ s *W > -• s jtion Noii.fiiiiSNi_ NviNosmiws^sa i ava a nf LI B RAR i eswsmithsonian jnstitutic ®v 2 i ^ ^ /tfgHB&s w ,V7 ~ O “ O XA(OSV^ 2 _J 2 nan libraries Smithsonian institution NoiiruiiSNi nvinoshiiins S3 i avua CO 2 m G3 » #*J OJ > l( 1/ ^3 Y5\ vgy m XQ\~5cjX ^ m co 2= co — co JTION NOimilJLSNI NVIN0SH1IINS S3U3VB8n LIBRARIES SMITHSONIAN INSTITUTK co z ... co 2 -w ^ 2 ,v* si . . .*i I ./ * "'1 q 1 v#k > 2 . _ '* Z aan2 UBRARIES^SMITHSONIAN^INSTITUTION^WOUnillSNI NVINOSHilWS^SB I a VB 3 CO 5 __ CO — , CO y^cnTN. a « i > w z \ w z < V § ^S§£§8\ < *. S ^8». < 2 H z ^s^\Vv^ -H 'wWy&s- 2 X ^ W O X UTION NOliniUShll NVIN0SH1IWS S3iavaan LIBRARIES SMITHSONIAN INSTITUTIC (o/ | h bfc: ^ g x>as^ . , itOliJ^i: n ^ m xayoTrtjy |j m AR I ESC,>SMITHS0NIAN”lNSTITUT!0NWM0lirUllSNl“NVIN0SHllWSS3 I avaail Ll B RAR 1 » z 0 ^ 2 ^ «g 2 > '' 2 'v<> > S X'.'ww'S^ > s msNi nvinoshiiwsotS3 i aVa a nzu b rar i esotsmithsonianZinstitution Nouruu “ V f/S “ m •»«' o -'^F' - o "•' 5 2 -* Z „J 2 ARIES SMITHSONIAN INSTITUTION NOIifUIISNI NVIN0SH11WS S3 1 d Vd 8 ll„ U B R AR z r- _ 2 r- ... 2 cs 33 > ^ 33 m S ^ UIISNI NVINOSHIIINS S3 I dVd 8 I “1 L 8 B R AR I ES^SMITHSONIAN~!NSTlTUTION NOIlfUU 2 5 CK‘* ^ £ < E t I 2 5 ' ^$0^ 5 2 > ^ 2 c n * -z — ___ S(/L -|- - UI1SNI-INVIN0SH11WSZS3 1 a Va 8 lA LI B R AR 1 ES*~ SMITHSON IAN-1 INSTITUTION NOIlfUU r- \. z r~ z <“ z H- _ g ‘ pi Xl'wKOjX 2 AR I ES^SMITHSONIAN- INSTITUTION^NOlinillSNI- NVIN0SH1IWS S3 I a Va 8 11 ~LI B RAR *AiV^ 2 < O in x s iuws^sa lava a i izli b r ar i esotsmithsonian“institution Nouruu (n — in =