Description of a new species of Caccoleptus ( Coleoptera: Dermestidae) from Brazil

Uma nova especie de dermestideo, Caccoleptus honeymani, e descrita e comparada com tres outras especies conhecidas. Somente existia uma outra especie sul americana, C. wicki da Colombia e Guiana. C. honeymani pode ser diferenciado deste pela presenca de um agrupamento de pubescencia branca na superficie dorsal e pela presenca de uma area pequena convexa triangular na parede anterior da fossa antenal na margem mediana. C. wicki nao tem pelos brancos no dorso mas na elitra tem duas faixas de pelo dourado. A parede anterior da fossa antenal da femea de C. wicki (o macho desta especie e desconhecida e e presumida reproduzisse partenogenicamente) e concava no seu comprimento total. Uma segunda especie de Entre Rios. perto do Rio de Janeiro, esta listada mais permanece sem nome devido ao pobre estado dos dois exemplares disponiveis. O macho desta especie e diferenciavel do macho de C. honeymani pela sua antena pectinada clavifomis. O clava da antena de C. honeymani tem os pedicelos dos segmentos perto do centro. Especies de Caccoleptus sao de interesse presentemente pelo seu possivel uso de controle biologico. C. wicki tem sido encontrado predando ovos e primeiros estagios larvais de Opisphanes cassina Fabr., uma lepitoptera ninfalida, defoliadora de bananeiras.

The genus Caccoleptus as presently re· cognized consists of 2 species from Central America and one species from Colombia, Guyana, and Panamá. Little is known of members of the genus except that larvae and adults of C. wicki Beal in Colombia are predators on eggs and first instar larvae of Opsiphanes cassina F., a nymphalid butterfly defoliator of banana plants (Beal, 1978). Shortly after I completed a review ot the genus in 1978, I found a few specimens of 2 additional species in material collected in Brazil by Herbert H. Smith and deposited in the collection of the Carne· gie Museum of Natural History. One of the 2 species, although represented by only 5 specimens, is being described at this time because of interest in members of the genus as possible agents ot biological control. I take pleasure in dedicating the species to Dr. Robert B. Honeyman, Jr., of San Juan Capistrano, California, a student of the histo· ry and literature of science. ches but patches ot elytra mostly on submedian and subapical areas. Head with integument dark reddish brown. Ratio of width o f interocular space to width across compound eyes 1:1 . 94. Punctures of vertex simple, equal in diameter to 1112 times diameter ot facet ot compound eye and separated by 1 to 2 times diameter ot single puncture. Antenna ochreous with 8-segmented club; segment 4 about 2 times as wide as segment 3; club not pectinate (pedicels inserted near center ot segments) . Pronotum with integument rubigi· naus except for suffused fuscous area on disc: punctures of disc simple with punctures equal in diameter to 1 to 11! 2 diameters of compound eye ; punctures separated by 1 to 3 times dia· meter of smaller punctures . Elytra with integument fuscous with suffused rubiginous maculae; maculae forming basal circle, short submedian band, and subapical band; submedian band close to and somewhat coalescing with posterior margin of basal circle. Thorac1c sterna with integument fuscous on sides oecoming rubiginous medially. Antennal fossa occupying ali of hypomeron except for small convexly rounded triangular area adjacent to compound eye; posterior margin enclosed entire length by knife-like carina. Prosternai process moderately broad, about 1 2/ 5 times as broad at levei of hind margin of front coxa as width of front tíbia. Mesosternum on each side of median sulcus somewhat oblique and about half as long as wide. Abdominal sterna with integument rubiginous. Sternum of (morphological) segment 8 as illustrated ( Fig. 2A).
Aedeagus and lateral lobes as illustrated (Fig.  2C); length of aedeagus not exceeding length of lateral lobes; base furcate with more or less square flange extending dorsad between lateral lobes from point of furcation; apex with minute proximally directed hook on dorsal side. Lateral lobes with relatively long apodemc at base; mediai extensions of lateral lobes (which in other species of Megatomini form a bridge between the lobes) present. not joined at midline but nea r midline curving toward ba· se. Legs ochreous. Ratio of width (measured across humeri) to length (of pronotum and elytra combined) 1:1 . 45. Length (of pronotum and elytra combined): 2.16 mm.

RANGE OF OBSERVED VARIATIONS -
Color of integument of pronotum and elytra as described above to entirely rubiginous except for suffused, slightly darker areas on elytra ~•t center of basal circles and at middle. Elytra with patches of white pubescence forming prc.minent subapical band but otherwise quite irregular. Ratio of length (of pronotum and elytra) to width (across humeri) varying from 1:1.35 to 1·1.45. Lengthe ranging from 2.16 mm to 2.52 mm. tural History that correspond with the museum accession number on the specimens indicate that they are part of a collection purchased from Herbert H. Smith i n 1906 and presumably collected by him (George Wallace, in litt ).
Paratypes deposited in the U. S. National Museum of Natural History, Instituto Nacional de Pesquisas da Amazonia, and the collection of the author.
Chapada, the type locality, is approximately 40 km (25 miJes) ENE of Cuiabá, the capital of the state of Mato Grosso. Collections de· signated with t his loca lity were taken by Smith both near the víllage at an elevation of 760 m and in tne surrounding regions to elevations as low as 550 m (Calvert, 1909). The real name of the village, according to Smith, is San· ta Anna la Chapada, but on current maps it is designated Chapada dos Guimarães. Smith collected here in 1885 and 1886 (Calvert, 1909). Santarém is on the amazon River at its junclion with the Tapajós River . Taperina is no doubt an alternative spelling for Taperinha, a plantation on the Amazon 32 km downstream from Santarém (Smith, 1879, p . 152). According to Holland (1919), Smith began his explorations of the Amazon in 1874 and conti nued collecting along its tributarias for the ensuing 3 years.
C. honeymani is readily distinguished from the only other described South American species, C. wicki, by the presence of spot-like clusters of white hairs on the elytra. C. wicki, known from the vicinity of Cúcuta in northeast Colombia and Georgetown, Guyana, is covered dorsally with light golden and dark brown to piceous hairs. The golden hairs of C. wicl<i form a broad submedian and an apical band. In addition, some specimens have a basal extension of the submedian band along the median suture. There may or may not be a short basal band. The irregular pattern of white patches of hair on C. honeymani contrasts strongly with this . C. honeymani may be distinguished from the 2 known Central American species, C. anisotomoides Sharp and C. rotundus Sharp, by the shape of the antennal fossa. In the Central American species the anterior wall of the fossa is concave its entire length, including the mediai margin next to the head. In C. honeymani there is a small but distinct triangular convexity on the anterior wall of the antennal fossa next to the head .

Caccoleptus species unnamed
The other Brazi lian species of Caccoleptus taken by Smith is represented by 2 specimen& from a locality designated as Entre Rios. The badly abraded condition of the specimens precludes their formal description at this time. The species is distinguished from C. honeymani and C. rotundus by the pectinate antennal club of the male, by the length of th aedeagus, Description of ... which exceeds the length of the lateral lobes, and by the lack of any suggestion of a bridge connecting the lateral lobes . Although there is more than one Brazilian locality named Entre Rios, this is said to be "on a branch of the Dom Pedro Segundo Railroad, where the latter meets the União e lndustria road" (Smith, 1879) . No locality named Entre Rios with such a description appears on current maps. The locality must, nevertheless, be in the general vicinity of Além Paraíba .
lt is worth noting that the lateral lobes of both C. honeymani and the unnamed species listed above lack a sclerotized median connection . Ali other species of Megatomini for which the genitalia have been examined. with the exception of some species of Anthrenus (e. g .. A. tadzhicus Mroczkowski, 1961), have a well-developed submedian bridge between the lobes. In C. honeymani (Fig . 2C) there is a partially formed bridge The unnamed species lacks any indication of a bridge (Fig . 2B) .
ACKNOWLEDGEMENTS I wish to acknowledge the generosity of George E. Wallace of the Section of lnsects and Spiders of the Carnegie Museum of Natural History for making these specimens avat· lable for my study and directing my attention to literatura describing the collecting ~ ~ti vities of H. H. Smith.