Louis M. Roth.
The Male Genitalia of Blattaria. I. Blaberus Spp. (Blaberidae: Blaberinae).
Psyche 76(3):217-250, 1969.

This article at Hindawi Publishing: https://doi.org/10.1155/1969/31394
CEC's scan of this article: http://psyche.entclub.org/pdf/76/76-217.pdf, 17552K
This landing page: http://psyche.entclub.org/76/76-217.html

The following unprocessed text is extracted automatically from the PDF file, and is likely to be both incomplete and full of errors. Please consult the PDF file for the complete article.

PSYCHE
Vol. 76 September, I 969 No. 3
THE MALE GENITALIA OF BLATTARIA.
I. BLABERUS SPP.
(BLABERIDAE : BLABERINAE)
BY Lours M. ROTH
Pioneering Research Laboratory
U. S. Army Natick Laboratories
Natick, Massachusetts 01 760
"Like many other genera the forms of which are variable and the specific features hard to ascertain and express, the genus Blaberus has been a despair to the systematist." (Rehn and Hebard, 1927). The use of male genitalia, specifically the aedeagus and preputial spines, has helped to alleviate some of the taxonomic difficulties of several species of Blaberus. Burmeister ( 1838 ; in Princis, 1946) first mentioned the preputial spines in Blaberus trapezoideus Bur- meister and Hebard ( I 91 7) described them in Blaberus craniifer Burmeister and B. atropos (Stoll). Princis (1946) illustrated the aedeagus and prepuce of the following species of Blaberus: giganteus (Linn.), trapezoideus, craniifer atropos, discoidalis Serville, para- bolicus Walker, anisitsi Brancsik, and bolh~iensis Princis. Lefeuvre ( 1960) illustrated the genitalia of craniifer, Quiaoit ( 1961) de- scribed them for craniifer and giganteus, and McKittrick (1964) illustrated discoidalis.
With the exception of Princis (1946) and Lefeuvre (1960), in- traspecific variations were not mentioned by the above workers. I have found considerable more variation in Blaberus genitalia than was indicated by Princis and Lefeuvre. In this paper I shall illus- trate the male genitalia of 12 species of Blaberus, describe group and specific differences, including intraspecific variations, and discuss the probable evolution of the aedeagus and prepuce in this genus. MATERIALS AND METHODS
The following 5 species of Blaberus were available in cultures:



================================================================================

218 Psyche [September
craniifer, giganteus, paraholicus, atropos, and discoidalis. In addition I have examined the genitalia of museum specimens of these species as well as those of B. co/osseus (Illiger), B. brasilianus Saussure, B. minor Saussure, B. fusiformis Walker, B. scutatus Saussure and Zehntner, B. anisitsi Brancsik, and B. boliviensis Princis. Of the 14
Blahus listed by Princis ( I 963) assellus (Thunb.) and latissimus (Herbst) were described from nymphs and are questionable species. For reasons given below, I consider Blaberus colosseus, which Hebard ( 1921 ) synonymized with B. giganteus, a valid species and B. trape- zoideus a synonym of B. craniifer.
As suggested by Princis (1946) the tips of the abdomens of dried specimens were dipped in hot water for about a minute, or the speci- mens were placed in a relaxing chamber. Once softened, the ab- domen was slit along the lateral membranes and the genitalia were removed usually without serious damage to the subgenital or supra anal plates. All specimens were treated with 10% KOH, cleared, and mounted in Permount. The hooked right phallomeres were mounted ventral side up and phallomeres LI and L2d were mounted dorsal side uppermost. The preparations of the prepuce were spread and flattened to show the spines. This should be taken into account when examining the illustrations. Normally the prepuce partly en- velopes L2d (see Fig. 121 in McKittrick's 1964 monograph which illustrates the folding of the prepuce in B. discoidalis). Although the principal genitalic characters used are L2d and the prepuce, I have also included photographs of R2 and LI for compara- tive purposes. Although these 2 phallomeres are very similar or have minor differences in all species of Blaberus (Figs. 1-24) they show family or subfamily differences and their inclusion should be useful in future studies of the genitalia of Blaberidae. Wherever known I have given locality data for the illustrated specimens, and the identity of the entomologist who determined the species. The abbreviations for the sources of this material are as follows (original geographical source, if known, follows the abbre- viations in the explanation of figures) : (N) = Natick culture ; (ANSP) = Academy Natural Sciences, Philadelphia; (IVICZ) =
Museum of Comparative Zoology, Harvard University; (L) = Zoological Institute, Lund, Sweden; (AMNH) = American Mu- seum of Natural History; (USNM) = United States National Museum ; ( BMNH) = British Museum (Natural History). Slides of genitalia are deposited with their respective males in the above museums.




================================================================================

19691 Roth - Genitalia of Blattaria 219
RESULTS AND DISCUSSION
The male genitalia of Blaberidae consist of 3 main structures ( McKittrick7s, I 964 terminology) . The right phallomere (R2) (Figs. 1-12) is a retractable hook, and all the species have a sub- apical incision (Fig. 2, arrow). The median sclerite (Lavm) is solidly attached (in Blaberus spp.) to L2d (L2 dorsal = the virga, penis, or aedeagus) (Fig. 52). The prepuce (mantle of Hebard's, 1917 terminology) is a soft, flexible membrane (Fig. 52) bearing characteristic spines, or truncate or rounded sclerotized elevations. The left side of the prepuce is solidly attached by sclerotization to the side of the L2 phallomere, whereas the right side is usually connected by a flexible membrane which permits it to fold partly around the virga.
One of the sclerites (LI) (Figs. 13-24) of the left phallomere in all the species of Blaberus studied are more or less similar and have a heavily sclerotized cleft, noted by McKittrick (1964).
Based on body size, color, and shape of the pronoturn, Hebard ( I 93 I ) placed fuxif ormis, brasilianus, anisitsi, and scutatzis in the Bra~ilian~us Group of the genus. Princis ( 1946) divided 8 species into the Giganteus and Atropos Groups, basing his divisions on the shape of L2d and the spines present on the prepuce. He did not examine the genitalia of minor, brasilianus, fusiformis, and scutatus, but suggested that the Brasilianus Group, established by Hebard, probably should be included in the Atropos Group. As a result of my examination of the 4 species not investigated by Princis, I believe Hebard was correct in erecting the Brasilianus Group and I place the above 4 species in this group.
Species of Blaberus can be readily b laced in their respective Groups, by the shape of the virga and preputial spines. However, variation is such that specific determination is often difficult. The 3 Blaberus Groups may be distinguished in the following key: I. L2d recurved dorsally and slightly to the right, usually forming a hooklike structure (Fig. 52). Extending dorsally and later- ally on the left, about where Lavm and Lad are solidly joined, is a sclerotized tumorlike outgrowth (Fig, 52, T). There are no large truncate elevations and all of the spines on the prepu- tial membrane are relatively small (Figs. 28 -40, 47-57, 62-74, 215 Giganteus Group.
L2d not hookshaped. Tumorlike outgrowth on the left side ab- sent. Prepuce with anterior truncate or rounded elevations almost always present on the left, and sometimes on the right side of the preputial membrane
2.




================================================================================

220 Psyche [September
Figs. 1-12. Right hooked phallamere (R2) of Blaberus spp. 1-2. GigaÌöteu Group. 1. B. crani~fer. (ANSP), Juxtiahuaca Cave, Colotlipa, Mexico (Jet. as B. irapezoidrv.~ by Rehn). 2. B. gigastev-s. (ANSP), Muzo, Colom- bia (arrow indicates the subapical incision). 3-6. bras ilia nu^ Group. 3. B.
scututux.
(ANSF), Ceara Mirim, Rio Grande do Norte, Brazil. 4. B. fu~i- formis. (ANSP), Provincia Sara, Dept Vera Cruz, Bolivia. 5. B, braat- lianus. (ANSI'), Natal, Brazil. 6. 5. minor. (ANSP) , Miiisian Tacaagk, Formosa, Argentina (det. Hebard). 7-10. Atropes Group. 7. B. atropos, (MCZ), Mexico. 8. 3. fafahvl'icus. (N), Ecuador. 9. B. discoidaff;. {N), Panama. 10. R. boSw'cn~i~. {L), Guayaquil, Ecuador (det. Princis). 11-12. Brua'ilianus Group. 11. B. cotosseus. (ANSP), Fyzabad, Trinidad. 12. 5. sp. D (probably ~olossfits). (ANSP), St. Laurent du Maroni, French Guiana (from specimen shown in Fig. 208). (all to scale shown in Fig. 12).



================================================================================

19691 Roth - Genitalia of Biattwia 22 I
Figs. 13-24, Left phallomere (Ll) of Maim spp. 13-14. Gtgantfvs Group, 13. 3. g'tganietts, (ANSP), Muzo, Colombia. 14. S. (raniijer (ANSP), Juxtlahuaca Cave, ColotSipa, Mexico (det. as B. trapemideus by Rehn). 15-18, 22. Brasitiunw Group. IS. B. brus'tl'iantis. (ANSP), Natal, Brazil. 16. B. scufahs. (ANSP), Ceara Mirim, Rio Grande do Norte, Brazil, 17. R. minor, (ANSP), Villa Ana F.C.S.F., Argentine Republic (det. Hebard). 18. S, fiiiformis. fANSP), Santa Cruz de 5a Sierra, Bolivia (det. Hebard). 19-21, 23, 24. Atropus Group. 19. B. faraffolicus. (N), Ecuador. 20. 3. discoidalis (N), Panama. 21. h'. atropus. (N), Trini- dad. 22. B. caiossw, (ANSP), Caparo, Trinidad (det. Hebard). 23, B. a~isitsi. (L), (det. Princis) . 24. B. bo/i&ensis. (L), Guayaquil, Ecuador (det. Princis). (all to scale shown in Fig. 18).



================================================================================

222 Psyche [September
Fig. 35. Upper. Tcgmen of B. craniifrr showing the distribution of the long setae on the marginal and acaptilar fie!& Lower. Anterior view of part of the tegmen showing the characteristic hairlike setae. 2. Anterior elevations usually rounded, present on the left and often on the right sides. Differences in size between anterior eleva- tions on the right and left sides not great. Preputial spines numerous, usually on the left and right sides and often occur in more than a single row (Figs. 76, 77, 79-82, 84-91, 93-1 I I, 21 1-214). In colossm the left preputial spines usually occur in a single row (Figs. 1 16-129) .. ,. Brasilianus Group. Truncate or rounded elevations usually present only on the left side and generally much larger and more robust than spines on the right. Preputial spines usually less numerous than in the Brasilianus Group, and are often arranged (when present) in a single row on the left, and single or sometimes double or ~artial double row on the right:. Spines on the right side usually more numerous than on the left (Figs. 133-153, 155-170, 174- 198, 200, 201, 203, 204, 210) .. . Atropos Group. Giganteus Group. - Two species of Blaberus hieanteus. Figs. 26, 27; craniifer, Figs. 41-46, 58-61) belong to this group. A useful tegminal character which Princis (1946) described can be used to distinguish B. craniifer from all other species of the genus. In
craniifer the marginal field and scapular field of the tegmina have diffuse projecting hairs (Fig, 25). According to Pincis, no other



================================================================================

19691 Roth - Genitalia of Blattaria 223
Figs. 25-44. 5. giganitus (Giganteus Gioup). 26. (N). 27. (ITSNM), St. Jean, French Guiana (det. as B. coloiseus by Hebard) (scale =z 10 mm). 28-40. L2d and prepuce (all to scale shown in Fig. 34). 28. (CSNM)
(from specimen shown in Fig. 27). 29. (USNM), Cablma, Panama. 30.
(USNM), Ft. Clayton, Canal Zone. 31. (AMNH), Barro Cobrado Island, Canal Zone. 32. (USNM), Chilibrillo Cave, Buenos Aires, Canal Zone. 33. (USNM), Puerto Berrio, Colombia, 34-35. (ANSP), Muzo, Colombia. 36. (USNM) , Atlantico, Colombia. 37. (AMNH) , Colombia. 38. ( AMNH) , Caripito, Venezuela. 39-40. (N) .




================================================================================

Figs. 41-57. B. rraniifa- (Gigan~us Group). 43-46. Adult males (scale 10 mm). 41. (USNM), Cordoba. Mexico (det, as B, trafmoidnus by Rehn). 42. (N). 43, (USNM), Rancho Qernado, Re. $5, Mexico. 44. (USNM), Teapa, Tabasco, Mexico. 45. (USNM), Key West, Florida. 46. (N), 47-57. L2d and prepuce (all to scale shown in Fig. 51). 47. (USNM) , Mexico (from specimen shown in Fig. 41). 48. (USMM), Mexico



================================================================================

19691 Roth - Genitalia of Blattaria 225
Blaberus has this character, although I have seen a few very minute hairs in some specimens of B. giganteus. In craniifer the tumorlike sclei-otized outgrowth on the left side of L2 extends caudally for a short distance and usually merges gradually into the border of the prepuce where small spines begin and form a fringe around the membrane. The sclerotized extension of the lateral outgrowth varies somewhat in length but is generally distinct (Figs. 47-51, 53-57, 62-64, 66-74). Exceptions are seen in Fig. 52 and 65. In B. giganteus there is usually little or no sclerotized extension from the tumorlike outgrowth into the preputial membrane, so' that the preputial spines begin more abruptly at the outgrowth (Figs. 28-35, 38-40). Exceptions to this are shown in Fig. 36 and 37. In both species the preputial spines may occur in more than a single row, and in some individuals there may be a reduction in the number of preputial spines (Figs. 37, 65, 70, 74). If Princis' (1946) tegminal character is valid for B. craniifer then I have not seen any correctly determined specimens of B. trapezoideus. All the specimens determined by Hebard or Rehn as tra?ezoide,us (Figs. 41, 58-61 ) have hairy tegmina and I therefore consider them to be light forms of B. craniifer. The genitalia of these "trapezoideus" (e.g., Figs. 47, 63, 64, 66) are indistinguishable from craniifer. Two specimens identified as B. trapezoideus, received from the University Zoological Museum, Copenhagen, Denmark were actually B. craniif er (Costa Rica) and B, parabolicus (Peru). According to Princis (1946) the preputial spines of trapezoideus are similar to giganteus but are smaller and more numerous, though always clearly separated from each other. Considering the variation in size, number, and spacing of preputial spines I doubt if this character can be used to distinguish trapezoideus from giganteus. Princis also (1958) states that the pronotum of trupezoideus is laterally truncated with approxi- mately parallel sides. Some individuals of our light-phased form of craniifer (in culture) (Fig. 42) also have the pronotum laterally truncate. It is possible that trapezoideus and craniifer are simply variants of the same species. The type localities of the 2 spp. of thme (from specimen shown in Fig. 43). 49. (USNM), Vera Cruz, Mexico. 50. (USNM), Tuxtepec, Oaxaca, Mexico (labelled trapezoideus) . 51. (USNM), Mexico (from specimen shown in Fig. 44). 52. (MCZ), Chichen Itza, Yuca- tan, Mexico. 53. (USNM), El Salvador. 54. (USNM), San Salvador, El Salvador. 55. (USNM), Benque Viejo, British Honduras. 56. (MCZ), Colon, Panama. (Abbreviations for Fig. 52. L2d dorsal left phallomere; L2vm = ventromedial left sclerite (fused to L2d) ; P = prepuce; T = tumorlike outgrowth.)




================================================================================

226 Psyche [September
FIRS. 58-74. B. cmnnfrr [Giganteus Group). 58-61. Adult malcq (scale = 10 mm). 5s. (ANSI'), Guatemala. 59. (ANSP), Vera Cruz, Mexico. 60. (ANSP). 61, (ANSP), Juxtlahuaca Cave, Colotlipa, Mexico. (These 4 males were determined as B. trupzoidcus by Rehn, or Hebard.) 62. (MCZ), Havana, Cuba (det. Rehn). 63. (USNM), Santiago-Vegas, Cuba. 64. (ANSP) (from specimen shown in Fig. 60). 65. (AMNH), Turrialba, Costa Rica. 66. (ANSP) (from specimen shown in Fig. 61). 67. (MCZ), Havana, Cuba. 68-69. (MCZ), Col6n, Panama. 70. (N). 71, (MCZ). 72. {ANSP), San Miguel, Vera Paz, Guatemala (det. as B. cola~si-us by Hebard). 73. (USNM), Florida (from specimen shown in Fig. 45). 74. (3).




================================================================================

19691 Roth - Genitalia of Blattaria 227
Gzganteus Group are : giganteus - "America" ; craniifer - Cuba. (see addendum regarding trafiezoideus) . Although color markings are variable in several species of Blaberus, they are especially so in B. craniifer. This species is represented by light (Figs. 41-43) and dark (Figs. 45, 46) forms with intermediates (Fig. 44) connecting the extremes. Markings of light phased individ- uals resemble B. giganteus (cf. Figs. 26, 27). Lefeuvre ( 1960) has described some color varieties which occurred in his laboratory culture of craniifer and we have cultures of both light and dark forms which crossed successfully. Lefeuvre claims that rearing craniifer for a number of years favored the formation of an "artificial subspecies'' which differed from the original in I) coloration of the pronoturn, 2) general coloration, in particular the male, and 3) the morphology of the penis and prepuce.
Lefeuvre suggested that the original B.
craniifer may have hybridized with a cIoseIy related species. The variations in craniifer which Lefeuvre described can be seen in museum specimens from different geog;-aphic localities. I have never seen any dark forms of B. giganteus comparable to dark craniifer. Brasilianus Group. - The 5 species in this Group, namely, scutatus (Fig. 75) ) brailianus (Fig. 781, fusiformis (Fig. 83), minor (Fig. 92), and colosseus (Figs. 112-1 15), show some genitalic differences in L2d and preputial spines but variation is so great within 4 of these species (Figs. 76, 77, 79-82, 84-91, 93-1 I I ) , that specific determina- tions, using genitalia alone, are often imposible. In some individuals of fusiformis (Figs, 85-87, 89, 91 ) and minor (Figs. 94-95> 100, 106) there is a marked reduction or loss of preputial spines and they may occur in a single row, usually on the left side. Characteristic of this group is the anterior elevations which are generally fused on the right side, and are about the same size as those on the left.
Rarely are the anterior spines on the right larger than those on the left (e.g., Figs. 84, 85, 88). The preputial spines de- crease only slightly in size from the anterior to posterior position. When the spines are numerous and occur in more than a single row they are often closely spaced and form a more or less dense uniform fringe around the preputial membrane (e.g.) Figs. 76, 77, 79-82, 88, 90) 93, 97, 99, 101) 105)
Hebard ( 1921, pa 148) stated) "From a study of the material in the Philadelphia collections, as well as specimens recently received from the Guianas, we are finally convinced that B. colosseus (Illiger) was based on a mere individual variation of giganteus, unworthy of nominal recognition." Princis ( I 963) followed Hebard and listed



================================================================================

Figs. 75-82. Bm~ilium~ Group. 75-77. B, scutatux, 75-76, ( ANSP), Pernambo, Brazil (paratype of 3. ~mtuha var. obscura S. and 2.). 77. (ANSP), Ceara Mirim, Rio Grande do Norte, Brazil. 78-82, B. .husi~ianus. 78-79. (ANSP), Natal, Brazil. SO. (ANSP), Independencia, Parahybq Brazil. gl. (MCZ), Brazil. 82. (USNM), Eatal, Brazil. (Scale for addm =z 10 mm; all genitalia to scale shown in Fig. 79.)



================================================================================

Figs. 83-91.
E. ju~ijurmi~ (Bradiunus Group). 83. Adult male (scale = 10 mm). (ANSP), Santa Cruz de la Sierra, Bolivia (det. Hebard). 84-91.
L2d and prepuce (all to scale shown in Fig. 85). 84. (ANSP), Carurnbo, Matto Grosso, Brazil (a portion of the prepuce on the right side is missing). 8.5. (ANSF), San Francisco, Argentina. 86. (ANSP), Provincia Sara, Dept. Vera
Cruz, Bolivia (det. Hebard).
87. (ANSP),
Bolivia (from specimen shown in Fig, 83). 88. (ANSP), Provincia Sara, Dept. Vera Cruz, Bolivia. g9. (ANSP), Jundiahy, Brazil. 90. (USNM), Wtiariti KO, Matto Grosso, Brazil. 91. (ANSP).



================================================================================

Psyche
[September
Figs. 92-107.
B. minor (Bradi~nus Group). 92. Adult male (scale = I0 mm). (ANSF), Argentina (det, Hebard). 93-107. L2d and prepuce (all to scale shown in Fig. 93). 93. (ANSPj, Paraguay. 94. (USNM), Natal, Brazil. 95. (USNM), Bmoklin, S5o Paula, Brazil, 96. (ANSP), Mission Tacaaglk, Formosa, Argentina. 97-98. (ANSF), Formosa, Argen- tina. 99. (ANSI'), Argentina (from specimen shown in Fig. 92). 100-102. (ANSP), Cham del Santiago del Ester0 Bords du Rio Selsdo Environs D'kaiio, Argentina. 103. { ANSP) , Formosa, Argentina. 104. (ANSF'). 105. (ANSP), Chaco de Santiago del Estero Rio Saiado, Argentina. 106-1 07. (ANSP) , Gran Chaco, Argentina.



================================================================================

19691 Roth - Genitaih of Bhttm-iu 23 1
Figs. 108-111.
Brmiliunw Group. Right and left sides of the prepuce. 108. 3. $cutatus (from Fig. 76). 109. B. bmdian~ds (from Fig. 79). Ilk B. minor (from Fig. 99). 111. 3. fu~ijormis (from Fig. 87). (scale = 0.1 mm).




================================================================================

232 Psyche [September
Figs. 112-122. B. colosseus (Brasilianuf Group). 112-1 15. Adult males (scale = 10 mm). 112. (ANSI'), Caparo, Trinidad. (This specimen is Fig. 4 in Hebard, 1916.) 113. (ANSP), Fyzabad, Trinidad. 114. (ANSP), Caparo, Trinidad, (This specimen is Fig. 5 in Hebard, 1916.) 115. {MCZ), Mexico. 116-122. L2d and prepuce (scale =: 0.2 mm). 116. (ANSP) (from specimen shown in Fig. 112). 117. (ANSP) (from specimen shown in Fig. 113). 118. (MCZ) (from specimen shown in Fig. 115). 31 19. (MCZ), Mexico. 120. (USNM), Trinidad, 121-122. (ANSP) (from specimen shown in Fig. 114; L2d and prepuce have been mounted separately).



================================================================================

19691 Roth - Genitaiia of Blattaria 233
Figs. 123-129, B. calosstus (Brasilianw Group). 123. (AMNH), Trini- dad. 1%. (USNM), Trinidad (some of the anterior elevations on the left side are broken off). 125. (AMNH), Rancho Grande, near Maracay, Vene- zuela. 126-129. Right and left sides of the prepuce. 126. (MCZ) (from Fig. 119). 127 (ANSP) (from Fig, 116). 128. (ANSP) (from Fig. 117). 129. (MCZ) (from Fig. US) (scale = 0.2 mm).



================================================================================

234 Psyche [September
Figs. 130-141. 3. atrapos (4fropss Group). 130-132. Adult males (scale == 10 mm). 130. (N), Trinidad. 131. (MCZ), Mexico. 132, (USNM), Colombia (taken in quarantine on bananas at Charleston, S.C.). 133-141. L2d and prepuce (all to scale shown in Fig, 140). 133, (MCZ) (from specimen shown in Fig. 131). 134. (USNM) (from specimen shown in Fig. 132). 135. (USNM), Georgetown, British Guiana, 136-138. (USNM), Trinidad. 139-141. (N), Trinidad.




================================================================================

Figs, 142-159. Prepuce of B. alrofin.
From Narick culture which ori-
ginated in TrSridad (ail to scale ahown in Pig. 142). - *
sA
4.
I




================================================================================

Psyche
-.
[September
Figs. IS+-165. 3. parubolicris (.'3iropos Group). 154. Adult male. (N), Puraquequara, Rio Negro, Amazonas, Brazil (scale = 10 mm). 155-165. L2d and prepuce (all to scale shown in Fig. 158). 155. (AMNH), Colombia, 156. (MCZ), Nap or Maranon (Ecuador Or Peru, northern Andes). (Type specimen of B!ab̤r armiym Scudder.) 157. (MCZ), Upper Amazon? 158. (USNM), Gaviio, Rio Negro, Amazonas, Brazil, 159. (N), Ecuador.



================================================================================

19691 Roth - Genitalia of Blattaria 23 7 Blaberus colosseus as a synonym of B. giganteus. Howevei-, Hebard (1916, p. 292) described the prepuce of co~ossez~s as follows: ". . . The surrounding mantle having the free dorsal and distal margins fringed with small blunt chitinous projections, these longer and more like short blunt teeth of a comb on the sinistral margin." This descriptioi~ does not fit a member of the Giganteus GI-oup. I have examined the male genitalia of several specimens including some which Hebard used in his study and conclude that coZosseus is not giganteus. The genitalia of coZosseus (Figs. I 16-129) differs ~nai-kedly from that of giganteus (Figs. 28-40). I?. colosseus phenotypically resembles gigan- teus, but it is paler in color (Hebard 1916) ; it is the largest member of the BrasiZimus Group, and the only species in this group which in size) color, and markings (some individuals) rese~nbles B. gigan- teus.
The prepuce of colosseu.~ combines features of both the BrasiZianus and Atropos Groups. The relatively small antei-ioi- elevations on the left side of the prepuce are not much larger than the spines on the right (Figs. I 16-120, I 22-129) a characteristic of the BrasiZianus Group. However, the preputial spines of cdosseus are all relatively large, fail-ly widely separated, particularly on the left side, and re- semble these spines in the Atropos Group. JVith few exceptions (Figs. 184, 188) species of the Atro$os Group have anterior truncate or rounded elevations on the left side of the prepuce that are ~nuch lasger and more robust than the spines on the right side (Figs. 133- 153, 155-170, 174-183, 185-187, 189-198, 200, 201, 203, 204, 210). One specimen from GuatemaJa, determined as B. colosseus by Hebard is actually BZaberus craniifer (Fig. 72 1. Two specimens (ex Canal Zone and French Guiana) determined by Hebard as colosseus are giganteus (Figs. 27, 28). The specimens which Hebai-d claimed were coZosseus came from Trinidad, Guatemala, Costa Rica, and Panama. I have seen 9 specimens of co/osseus; 6 were from Trinidad, 2 from ILIexico, and I from Venezuela. One specimen from French Guiana is probably colosseus (Figs. 208, 214). The distribution of this species must await an examination of additional material. Atropos Group. - Five species, atropos (Figs. I 30- I 321, para- bolicus (Fig. I 541, discoidalis (Fig. I 73) ) boliviensis (Fig. I 991, and anisitsi (Fig. 202) belong to this group. The armament on the PI-eputial membrane shows the greatest vai-ia- 160. (N) , Borba, Rio Madeira, Amazonas, Brazil. 161. (AMNH) , Iquitos, Peru. 162. (AMNH), Rio Ucayali, Peru. 163. (AMNH), Moyobamoa, San Martin, Peru. 164. (AMNH), Rio Ucayali, Peru. 165. (AMNH), Rio Marafion, Peru.




================================================================================

Psyche
[September
Figs. 166-172.
B. parabolicus (Afropo~ Group), Ud and prepuce- The two sttuctures have been mounted separately in Figs. 168-172 (all to sc& shown in Fig. 169). 166-167. (USNM), South America. 163-172. (N}, Puraquequarz, Rio Negro, Amazonas, Brazil.



================================================================================

Figs. 173-IW B. discoidah (Atropos Group), 173. Adult male {scale =Z 10 mm). (N), Panama. 17+188. L2d and prepuce (all to scale shown in Fig. 177). 174-175. (MCZ), Nicaragua. 176-177. (N), Panama. 178, (USNM}, Pontarenas, Costa Rica. 179. (MCZ), Trinidad. 180. {AMNH), Colombia. lg1. (USNM), Trinidad. 182. (MCZ}, Panama. 183. (AMNH), Barro Colorado, Canal Zone, Panama. 184. {USNM), Colombia (from wild orchids at Hoboken Quarantine). 185. (AMNH), Barro Colorado, Canal fine, Panama. 186. (USNM), British Guiana, 187. (N), Panama. lE8. (AMNH), Puerto Plata, Dominican Republic.



================================================================================

240 l'xychc [September
Figs. 189-198. B. diacoido!i~ (Atropos Group). L2d and prepuce; in 194-198, L2d has been removed (all to scale shown in Fig. 196). 189. (USNM), Fhiador. 190. (USNM), Venezuela. 191. (AMNH), Colombia. 192. (AMNH), labeled "Africa" which is undoubtedly an error, E93-197. (N), Panama. 198, (MCZ).




================================================================================

Figs. 199-204. Blabcrux spp. (A~~o~oJ Group}. Genitalia from the adult specimens shown. 199-201. B. hiiwiensis. (L) , Guayaquil, Ecuador (det. Princis). 202-204. B, afiirif~i. (L) (det. Princis). Fig, 201 and 204 are enlargements of the preputial spines of the specimens shown in Fig. 200 and 203 (scale for adults = 10 mm, for genitalia = 0.2 mm).



================================================================================

242 Psyche [September
tion in atropos (Figs. I 33-1 53 ) . Princis ( I 946) stated that the trun- cate elevation on the left side was 3-pronged whereas Hebard ( 1917) stated it was 2-pronged. These elevations may vary from a single arm (Fig. I 52) to one consisting of more than 5 prongs (Figs. 133, 134). More striking is the marked reduction in numbers of the smaller pi-eputial spines in some males. This usually occurs on the left side (Figs. 139-141, 144-153) where there are few spines to begin with but a marked reduction may occur even on the right side (Figs. 141, 153). In some males the spines are completely absent from the left side (Figs. 140, 141, 147, 149-153).
In parabolicus (Figs. 155-170) most of the spines on the right side of the prepuce are more or less pointed. In some males there is a reduction in number of spines (Figs. 156, 169). Bruijning (1959) described the genitalia of parabolicus as follows: "At the extreme right of the preputium stout, rounded processes are inserted between the teeth; some of the teeth on the free margin form pairs which are squarely inserted on the margin; sinistsad the teeth are developed in blunt, stout, chitinous processes, while at the extreme left a large bi- to trilobate process is found . . ." An examination of Figs. 155-170 shows that variability is so great that it is impossible to indicate specifically the number of truncate elevations on the left side or the exact arrangement and shapes of the preputial spines on the right. In B. discoidalis (Figs. I 74-1 98) the truncate elevations on the left usually arise very close to Lmm, extend dorsally, and may even overlap Lavm (Figs. I 75-180, 190). Preputial spines are more nu- merous on the right side, usually are somewhat truncate but some- times are rounded or pointed at the tips. In some specimens (Figs. 195, 198) the right anterior spines, though smaller, tend to resemble the large elevations of the left side except that they are rarely fused at their bases (Fig. 195). The spines decrease in size distally on the membrane and in some individuals there is a reduction in number, usually on the left side.
Lad is variable in size and shape. In a few males the large truncate elevations which are highly variable in num- ber on the left side are poorly defined or fused together (Figs. 182, 184, 188, 191 ) and sometimes (Fig. 184) are reminiscent of the tumorlike outgrowth on the left side in the Giganteus Group. According to Princis (1946) the preputial armament is simple and sparse in anisitsi (Figs. 203, 204). His drawing shows a 3-pronged truncate elevation plus 3 bluntly rounded spines on the left side and only 5 smaller pointed spines on the right. The specimen shown in Fig. 203 was the one used by Princis (Fig. 6 in 1946; and identified



================================================================================

19691 R 0th - Genitalia of Blattaria 243 by him with reservations) but he did not remove the genitalia from the male; it differs from Princis' description in having a 4-pronged elevation on the left and more spines on the right (than he figured) terminating anteriorly in 3 spines fused at their bases (Fig. 204). Hebard stated that the species related to B. fusiformis Walker . . . are poorly understood and the description of fusiformis is vague. If our specimens are correctly determined it is possible that anisitsi is a synonym, based on material showing decided depauperation." Although Hebai-d included anisitsi in the Brasilianus Group, the specimen identified by Princis as anisits'i (Fig. 202) is clearly a mem- ber (by male genitalia, Fig. 203) of the Atropos Group (Princis 1946). If both Princis and Hebard are correct in their determina- tions anisitsi and fusiformis are obviously not the same species. Pi-incis (1946) compared the genitalia of boliviensis (Figs. 200, 201 ) with anisitsi (Figs. 203, 204). According to him the penis in boliviensis is more massive. The spines on the right side start with a bluntly rounded spine, are larger, more numerous and not as widely separated as in anisitsi. On the left side there is a 3-pronged truncate elevation followed by 10 (according to his drawing) truncate or rounded spines set fairly close together. In the specimen shown in Figs. 200, 201 (not the one illustrated by Princis), the truncate ele- vations on the left are at least 6-pronged. No doubt an examination of additional specimens of these 2 species would show as much intra- specific variation as occurs in other species of the Atropos Group. The truncate elevations on the left side of both anisitsi and bolivien- sis arise close to Lzvm, extend dorsally, and their genitalia closely resemble those of B. discoidalis.
Undetermined species. - Several museum specimens were examined whose genitalia and phenotypic appearance did not fall into the known species. These were as follows:
1-2. Blaberus spp. A (Figs. 205, 213) and B (Figs. 206, 211, 2 I 2). - These 2 species from Peru, except for their much smaller size, resemble colosseus, particularly in their slender form and pale coloration. The pi-eputial spines of both forms (cf. Figs. 21 I, 213) differ from each other. The anterior elevations of the left side are not much larger than those on the right, thus resembling the prepuce of colosseus.
3. Blaberus sp. C (Fig, 207). - This specimen from Colombia was identified by Hebard as B. discoidalis. However, it is considerably smaller and more slender than is discoidalis, and phenotypically re- sembles the specimen identified by Princis as B. anisitsi (Fig. 202). Its genitalia (Fig. 210)
are unique (and differ from anisitsi, Fig.



================================================================================

244 Psyche
[September
Figs. 205-215.
Males of Blaberus spp. The genitalia are from the adult males shown. 205, 213. Blaberus sp. A. (Brasiliwus Group) (USNM), Tingo, Maria, Peru. 206, 211, 212. Blabcrus sp. 3. (Brusilianus Group) (USNM), Tingo, Maria, Peru.
(Fig. 212 is an enlargement of the right and left sides of the prepuce shown in Fig. 211.) 207, 210. Blz~bervs sp. C. (Atrapos Group) (USNM), Susumuco, Colombia. (The prepuce in Fig. 210 is broken and the spines on the lower left side normally lie under L2d.)
(det. by Hebard as B. dhcoiduh).
208, 214. Rlaberus sp. D. (Bra-
siiiavus Group) (AMSP), St, Laurent du Maroni, French Guiana (det. by Hebard as B. gigunfeus}. 209, 215. Bluwerus sp, E. (Gigaateui Group) (USNM), Borba, Rio Madeira, Amazonas, Brazil. (Scale for adults = 10 mm; for genitalia := 0.3 mm.)




================================================================================

19691 Roth - Genitalia of Blattaria 245
Fig. 216. Prepuce of Blabetus sp. (Gigantexts Group). (MCZ), Ande- goya, Colombia. Different portions (brackets) of the prepuce (center) are enlarged to show variations in the spines (scale = 0.1 mm).



================================================================================

Psyche [September
Fig5 2 17-220. Klabrrvs ~rafezoidtus. Pronoturn and gcilitalia. 217-21 S. (BMNH). ParaIectotype of Shhcru gitadrifrra Walker. 219-220. (BMNH). Lectotype of Blabera quadrifrra Walker. Oajaca, Vera Cruz, Mexico. 221- 224. fihhcrus crmiifer. Variations in the shape of the pronoturn in speci-



================================================================================

19691 Roth - Genitalia of Blattaria 247
203) in that all of the spines bordering the prepuce are relatively large and arise from a well defined sclerotized margin; the shapes of the preputial spines also differ between these 2 specimens. 4. Blaberus sp. D (Fig. 208). - This is a species determined by Hebard as B. gigantpus (from French Guiana) but its genitalia (Fig. 214)
are that of a member of the Brasilianus Group. The truncate elevations on the left side of the prepuce are small and some- what like those of colosseus, but the other spines are greatly reduced in size and number. This specimen is probably colosseus (though it is somewhat broader and more intensely colored than colosseus from Trinidad and Mexico,
(cf. Figs. I I 2- I I 5 ) in which the preputial spines have been greatly reduced (cf. Figs. I 16-129). 5. Blaberus sp. E (Fig, 209). - This specimen is close to gigan- teas but is more slender and its general coloration is very pale. Its genitalia (Fig. 215) are massive and there are more rows of prepu- tial spines than are usually found in giganteus (Figs. 28-40). Distribution. - The species of Blaberus are almost entirely neo- tropical (Table I). Four of the 5 species of the Brasilianus Group are restricted to South America; colosseus is more widely distributed and occurs in Mexico, Central and South America. Members of the Giganteus and Atropos Groups are found in Central and South America, and a few species occur in the West Indies, southern Florida, and Mexico.
Evolution of the aedeap-us and prepuce. - I believe that the prepuce of Giganteus Group males which lack truncate elevations and have relatively simple, small preputial spines is the most primitive of the 3 groups of Blaberus. However, though the preputial spines are small they may vary in shape (Fig. 216) and some spines are rem- iniscent of those found in the Brasi/ianus and L4tropos Groups. A Giganteus Grouplike form could have given rise to individuals of both the other 2 groups. The preputial spines of the Brasilianus Group are often numerous and may occur in multiple rows (e.g., Figs. 88, 90) like some individuals of the Giganteus Group (e.g., Figs. 36, 47, 215). In both the Brasilianus and Atropos Groups, the anterior elevations of the prepuce on the left side probably evolved from the left tumorlike outgrowth of a Giganteus Grouplike form. However, in the Brasilianus Group, the anterior elevations on the left and right sides do not differ greatly in size whereas there is a marked size difference between the elevations on the two sides in the mens from a laboratory culture. (Scale for pronoturn [see Fig. 2241 :== 5 mm.; scale for genitalia [see Fig. 2181 = 0.5 mm).



================================================================================

248 Psyche
[September
Atropos Group. It is of interest that in occasional specimens of B. discoidalis (Atropos Group) the anterior elevations on the left side are so poorly developed (Fig. 184) that the lateral swelling of L2d resembles the outgrowth in the Giganteus Group. In Princis'
( 1963) linear arrangement of 14 Blaberns spp., the species minor is separated from other members of the Brasilianus Group by boliviensis and atropos.
I would rearrange this sequence
and place minor with brasilianus, fusiformis, and scutatus. Table 1. Geographical distribution of species of B'aberus. --
1 Distributions
Species
Giganteus Group
craniifer
giganteus
Brasilianus Group
brasilianus
colosseus~
fusiformis
minor
scutatus
Atropos Group
anisitsi
atropos
boliviensis
discoidalis
parabolicus
Mexico, Guatemala, British Honduras, Honduras, Costa Rica, Panama, Venezuela, Cuba, Dominican Republic, Florida (Key West)
Mexico, Guatemala, Panama, Colombia, Venezuela, Trinidad, British Guiana, Surinam, French Guiana, Dominican Republic ( ?)
Brazil
Trinidadb, Mexicob, Guatemala, Costa Rica, Panama, Venezuela^, French Guiana
Brazil, Bolivia, Paraguay, Argentina
Brazil, Bolivia, Paraguay, Argentina
Brazil, Peru
Paraguay
Trinidad, British Guiana, Chile ( ?), Colombiab, Mexicob
Bolivia, Ecuadord
Costa Ricab, Dominican Republicb, Jamaica, Cuba, Haiti, Vieques Island, Panama, Colombia, Venezuela, Trinidad, Ecuadore, Nicaragua^, Florida (near Key West) f
Colombia, Surinam, Brazil, Peru, Ecuador, Bolivia aFrom Princis (1963) unless otherwise indicated; the localities for craniifcr include those for trapezoidcus.
^From present study.
localities from Hebard (1920).
dFrom Princis (1952).
~Princis lists Ecuador with a ?. I have seen 1 specimen (Fig. 189) from Ecuador.
fFrom A. B. Gurney (~ersonal communication). (Record of USNM and Fla. Plant Board).




================================================================================

19691 Roth - Genitalia of Blattaria 249
Chromosome numbers. -The diploid chromosome numbers of fe- males of 5 species of Blaberus are: giganteus, 74; craniifer, 74; atropos, 74 ; parabolicus, 40; discoidalis, 38. The males have one less sex chromosome (Cohen and Roth, unpublished data). Evidently members of the Atropos Group have variable chromosome numbers. Addendum: Since this paper went to press I have examined the lectotype and paralectotype of Blabera quadrifera Walker, which is a synonym of B. trapezoidws Burm. The pronotum of one of these specimens is illustrated by Princis ( 1958, p. 74). The tegmina of these 2 specimens are hairy, as they are in B. craniifer, and their genitalia (Figs. 218, 220) are indistinguishable from those of craniifer (cf. Figs. 47-57, 62-74). As for the laterally truncate pronotum of trapezoideus (Figs. 217, 219), the pronotal shape is so variable in light colored craniifer (Figs. 221-224) that this character cannot be used to distinguish the 2 species. I believe that trapezoideus (type locality Mexico) is the light colored form of craniifer. Both species were described by Burmeister (Handb. Ent. 2 (2), Berlin, I 838, p. 5 I 6). Because B. craniifer has been used widely as an experimental animal I select it as the valid name for this species. Based on the structure of the prepuce and aedeagus, species of Blaberus are placed in the following three groups: I) Giganteus Group (giganteus, craniifer), 2) Brasilianus Group (minor, brasili- anus, colosseus, fusiformis, scutatus) , and 3) Atropos Group (atropos, parabolicus, discoidalis, anisitsi, and boliviensis) . I consider Blaberus trapezoideus to be a synonym of B. craniifer, and B, colosseus, formerly a synonym of B. giganteus, to be a valid species.
The genitalia are sufficiently distinctive to place individuals in their respective groups. However, intraspecific variation of the genitalia is so great that it is difficult and sometimes impossible to distinguish between certain species of a Group.
Three of the 5 species in the Atropos Group have variable chromo- some numbers of 74, 40, and 38.
ACKNOWLEDGEMENTS
I thank the following for the loan of Museum material: Dr. E. M. Emsley, Academy of Natural Sciences, Philadelphia, Dr. A. B. Gurney, U. S. National Museum, Dr. Howard Evans, Museum of Comparative Zoology, Harvard University, Dr. Karl Princis, Zoolog- ical Institution, Lund University, Sweden, Dr. Jerome G. Rozen, Jr.,



================================================================================

250 Psyche [September
American Museum of Natural History, Dr. S. L. Tuxen, Zoological Museum, Copenhagen and Dr. David R. Ragge, British Museum (Natural History), London. I am indebted to Miss Johanna Darl- ington, University of West Indies, Trinidad, who sent me living atropos from which our culture was started. I collected live speci- mens of B. parabolicus and other species (preserved) during Phase C of the Alpha Helix expedition to the Amazon in 1967. I thank the National Science Foundation for support on the Amazon expedition under Grant NSF-GB-5916. I am grateful to Mr. Samuel Cohen for taking the photographs.
REFERENCES
BRUIJNING, C. F. A.
1959. The Blattidae of Surinam. Studies of the fauna of Suriname and other Guyanas. Vol. 2: 1-103.
HEBARD, M.
1916.
Critical notes on certain species of the genus Blaberus (Orthop- tera, Blattidae) . Entomol. News, 27 : 289-96. 1917. The Blattidae of North America, north of the Mexican boundary. Mem. Amer. Entomol. Soc. No. 2: 1-284.
1920. The Blattidae of Panama. Mem. Amer. Entomol. Soc. No. 4: 1-148 (1919).
1921. Studies in the Dermaptera and Orthoptera of Colombia. Second paper. Dermaptera and orthopterous families Blattidae, Mantidae and Phasmidae. Trans. Amer. Entomol. Soc. 47: 107-169. 1931. Die Ausbeute der deutschen Chaco-Expedition 1925/26. - Orthop- tera. Konowia, 10 : 257-285.
LEFEUVRE, J. C.
1960. A propos de Blabera craniifer Burmeister 1838 (Insecte dicty- optere). Bull. Soc. Scient. Bretagne, 35 : 145-161. MCKITTRICK, F. A.
1964. Evolutionary studies of cockroaches. Cornell Univ. Agric. Exp. Sta. Memoir 389: 1-197.
PRINCIS, K.
1946. Zur Kenntnis der Gattung Blaberus Serv. (Blatt.) . Opusc. Entomol. 11 : 139-146.
1952. Reports of the Lund University Chile Expedition 1948-1949. 8. Blattariae. Kungl. Fysiogr. Sallsk. Handling. 63: 1-11. 1958. Revision der Walkerschen und Kirbyschen Blattarientypen im British Museum of Natural History, London 11. Opusc. entomol. 23: 59-75.
1963. Orthopterorum Catalogus. Blattariae. Pars 4 : 76-172. s'-Gra- venhage.
QUIAOIT, E. R.
1961. An investigation of growth, development and dimorphism in cockroaches (Orthoptera : Blattidae) . Doctoral Dissertation, Kansas State University.
REHN, J. A. G. and M. HEBARD
1927. The Orthoptera of the West Indies. Number 1. Blattidae. Bull. Amer. Mus. Nat. Hist. 54: 1-320.




================================================================================