Ann. Rev. Entomol 1979. 24:449-73 Copyright @ 1979 by Annual Reviews Inc. All rights reserved BIOLOGY OF THE BRUCHIDAE +6178 B. J. Southgate Biology Department, Pest Infestation Control Laboratory, Ministry of Agriculture, Fisheries, and Food, Slough SL3 7HJ, Berks, England INTRODUCTION Species of Bruchidae breed in every continent except Antarctica. The larg­ est number of species live in the tropical regions of Asia, Africa, and Central and South America. Many species have obvious economic importance because they breed on grain legumes and consume valuable proteins that would otherwise be eaten by man. Other species, however, destroy seeds of an immense number of leguminous trees and shrubs, which, though they have no obvious economic value, stem the advance of the deserts into the marginal cultivated areas of the world. When this ecosystem is mismanaged by practices such as over­ grazing, then any organism that restricts the normal regeneration of seed­ lings will, in the long run, affect agriculture adversely. This has been demonstrated recently in some African and Middle Eastern semiarid zones (65). The present interest in the management of arid areas and in the introduc­ Annu. Rev. Entomol. 1979.24:449-473. Downloaded from www.annualreviews.org Access provided by Copyright Clearance Center on 11/01/20. For personal use only. tion of alternative tree species to provide timber, fodder, or shade has stimulated a detailed study of the ecology of some leguminous trees and shrubs that has revealed some deleterious effects of bruchid beetles on the seeds of these plants (42, 43, 59). It has also emphasized the inadequacy of our knowledge of the taxonomy and biology of these beetles. Simulta­ neously there has been a move to improve grain legume crops by breeding varieties that give heavier yields and are resistant to the pests and diseases that devastate current varieties. Thus ecologists, biochemists, agronomists, and plant breeders are already cooperating, and I hope this review will encourage entomologists to study the Bruchidae to gain a greater under­ standing of their ecology. 449 0066-4170/79/0101-0449$01.00 450 SOUTHGATE THE TAXONOMIC STRUCTURE OF THE FAMILY BRUCHIDAE The family Bruchidae at present consists of approximately 1300 species, grouped into 56 genera placed within 5 subfamilies. It is placed by most authorities in the superfamily Chrysomeloidea, and although some detect affinities with the Curculionoidea, Crowson (31), in particular, argues against that view. It comprises five subfamilies, Amblycerinae, Bruchinae, Eubaptinae, Kytorhininae, and Pachymerinae, as definedby Bridwell (15). To these, Luk'yanovich & Ter-Minasyan (68) add the subfamily Rhaebinae, represented by a single genus Rhaebus. Bridwell had excluded this although he had seen no specimens of it. In my opinion it fits into the Bruchidae, although it has some affinities with the Chrysomelidae. Its larvae develop within fruiting bodies, and although not of taxonomic significance, it is similar to the Bruchidae. Bottimer (10) accepted its exclusion, but Crowson (31) notes that it has close affinities with the Bruchidae and infers that it should be placed in the family. From the time of Linneaus up to the early 1900s, interest in the Bru­ chidae was limited to the description of considerable numbers of new spe­ cies, which were usually designated Bruchus. Bridwell (16), who made the first serious study of classification, constructed a key based on adult mor­ phology to the genera of bruchids found in America north of Mexico. Pfaffenberger& Johnson (75) made a limited generic classification based on larval morphology, especially on the presence or absence of ocelli and on the development of legs in the mature larvae, but they made no attempt to show any biological significance in these structures. At species level, the New World species have been extensively studied. Johnson and Kingsolver and co-workers (53, 56, 58, 100) have each as­ signed numerous species to genera that had been designated Bruchus by early workers such as Sharp. Less has been done on Old World genera such as Bruchidius, but Decelle (33) has separated and defined Conicobruchus Annu. Rev. Entomol. 1979.24:449-473. Downloaded from www.annualreviews.org and Tuberculobruchus. Bruchidius remains a heterogeneous group defined Access provided by Copyright Clearance Center on 11/01/20. For personal use only. by Schilsky (84) as having a conical thorax and a single spine on each hind femora. This spine may be almost invisible, or approximately 0.2 mm long. Species of the genus are found in Europe, Asia, and Australia, but King­ solver (62) evidently believes that the Australian species should be placed in some other genera. In my opinion the remaining African species, too, should be reassigned. A problem that arises when active taxonomic studies are in progress, especially in groups that have some economic importance, is the continued use of obsolete names. Applied entomologists are often unaware of new synonymies and of changed opinions about generic status, and names are THE BRUCHIDAE 4S 1 changed after papers are published. Therefore, in this paper I place the original name in parentheses after the currently accepted generic name whenever a species is mentioned, for example. Caryedon (=Pachymerus) interstinctus. Biotaxonomy The current interest of biochemists in the chemical composition of legume seeds has provided information that can be correlated with infestation and can generate ideas on the phylogenetic origins of the Bruchidae. Foodstuffs, ecology, and behavior may be adaptive, but they may also have some taxonomic value if they cause divergence of forms. Thus the inclusion of biochemical data about the seeds of many legumes should help the ecologist understand why some seeds are vulnerable to bruchids and how others combat attack, and perhaps gain insight into the evolution of species. Interesting associations occur between host plant groups and bru­ chid subfamilies. Thus the Bruchinae for the most part are associated with the family Leguminosae. whereas the Amblycerinae mostly attack other families (Table 1). The Eubaptineae and Pachymerinae attack mostly non­ leguminous plants. Host specificity of itself is not a valid criterion on which to base taxonomic differentiation. but it is a useful addition to the more conventional methods. Johnson (54) has suggested affinitiesbetween species of Acanthoscelides and Abutiloneus idoneus because they both breed in the Malvaceae. Likewise his analysis of the host relationship of the Acanthosce­ !ides produces some groupings that would not be made from a purely morphological analysis. Janzen (48) suggests that host species divergence is based less on physical characteristics. such as the seed type, pod structure, and shape or nature of the seed coat, than on nutritional characteristics of the seed. This hypothesis has been examined only for a few New World genera and could well be extended to the Old World. Annu. Rev. Entomol. 1979.24:449-473. Downloaded from www.annualreviews.org Access provided by Copyright Clearance Center on 11/01/20. For personal use only. GEOGRAPHICAL DISTRIBUTION The early bruchid taxonomists lumped all the species into a few large genera, so no distribution pattern could be discerned. Now that our in­ creased knowledge enables us to differentiate the species into more distinct genera, it is evident that except for the very few species dispersed by commerce, almost all the genera are limited to either the Old or the New World. Bottimer (10) admirably defines the five subfamilies comprising the Bru­ chidae, along with the seven tribes into which two of the subfamilies are split. About three quarters of the genera are still placed in the subfamily 452 SOUTHGATE Table 1 Plant families other than Leguminosae attacked by Bruchidaca Bruchidae Subfamily Tribe Genus Plant family attacked / Amblycerinae Amh/ycerus Anacardiaceae (51) Boraginaceae (51, 102) Euphorbiaceae (51) Malpighiaceae (51) Malvaceae (51) Nyctaginaceaeb Sterculiaceae (55) Tileaceae (61) Verbeneaceae (61) Vitaceae (57) Spermophagus Boraginaceae (102) Convolvulaceae (2, 45, 102) Malpighiaceae (102) Malvaceae (88, 102) Bruchinae Acanthoscelidini Ahutiloneus Malvaceae (16) Acanthoscelides Anacardiaceae (18) Cistaceae (54) Lythraceae (54) Malvaceae (25,54) Onagraceae (54) Rhamnaceae (54) Sterculiaceae (55) Tiliaceae (51) Althaeus Malvaceae (16) Bonaerius Malvaceae (17) Caryedes Euphorbiaceae (102) Cosmobruchus Asteraceae (14) Dahlibruchus Asteraceae (14) Lithraeus Anacardiaceae (18) Neltumius Rhamnaceae (25) Stator Bixaceae (52) Myrtaceae (52) Bruchidius CistaceaeC Annu. Rev. Entomol. 1979.24:449-473. Downloaded from www.annualreviews.org Megacerini Megacerus Convolvulaceae (25) Access provided by Copyright Clearance Center on 11/01/20. For personal use only. Eubaptineae Eubaptus Acanthaceae (95) Kytorhinineae Kytorhinus Zygophyllaceae (2) Pachymerinae Caryedon tini Caryedon Apiaceae (� Umbeliferae) (2,89) Combretaceae (78) Pachymerini Pachymerus Palmae (13, 102) Pandanaceae (13) Caryoborus Palmae (13, 102) Caryobruchus Palmae (13) Caryotrypes Pandanaceae (36) a Numbers in parentheses indicate reference. bOata on specimen in the British Museum of Natural History. C B. J. Southgate, unpublished information. THE BRUCHIDAE 453 Bruchinae, which is split into four tribes. Of these, two are exclusively Old World origin, one is exclusively from the New World, and one very large New World tribe, the Acanthoscelidini, has a few Old World species in one genus,