Characters that are referred to as of general comprehension are those proposed several times in the literature as being important to species definition, but were actually found to be more generalized, i. e., they in fact occur in higher taxa. General characters are found in all, or most, male of diabroticites analyzed, and might also support the definition of larger taxonomic groups: 1) Smaller body size: considered general for Chrysomelidae, observed for most diabroticites; 2) Bigger eyes (relative to the total size of the head): cited often in species descriptions, but actually observed in most diabroticites; 3) Tarsal adhesive disks: structures present in most Chrysomelidae, with variation found among subfamilies and often among tribes, regarding the number of legs in which they occur and the proportion of the dorsal surface that they occupy (Stork 1980). For Diabroticites, the adhesive disks are present at the first tarsomere in pro- and mesothoracic legs; 4) Emargination on the posterior margin of ventrite V: this characters sometimes supports the definition of the subfamily Galerucinae, and its shape can be diagnostic of tribes, being rounded in Luperini and Galerucini, for example (Bechyné and Bechyné 1962, Wilcox 1965). In diabroticites this emargination is, usually, round; and 5) Prothoracic tibiae with continuous apex, without spurs: probably constant in the whole tribe (Wilcox 1965).

Special characters are those used to describe lower taxonomic ranks, i. e., genera and species, and have been or not mentioned as diagnostic features in original descriptions.

With the exception of two genera, Amphelasma Barber, 1947 and Cochabamba all genera in Diabroticites are represented with special dimorphic characters. Of all species analyzed, only 47 were found to have special dimorphic characteristics. This is interesting, since the original definitions of diabroticites genera were usually based in sexual dimorphism. In accordance with the pattern observed for the subfamily (Mohamedsaid and Furth 2011), special dimorphism was observed in every tagma in the analyzed taxa of Diabroticites (Table 3).

Out of the 12 genera recognized by dimorphic characters, 4 are monotypic (Buckibrotica, Palmaria, Platybrotica and Pseudodiabrotica). The diagnostic characters and its validity will be discussed for each tagma.

Most of the dimorphic characters were found in the head, mainly in the antennae, a proportion which agrees with the general pattern observed in the subfamily (Mohamedsaid and Furth 2011).

Gynandrobrotica was described as showing an “excavated clypeus”, without any further details or illustrations. All four species studied have the same kind of modification, which is better described as the frons being elongated and with a shallow, smooth, round concavity (Figure 1), accompanied by sparse, large punctuation. It is interesting to compare this character with the differently excavated type of frons found in species of Cerotoma Chevrolat, 1837 (Figure 2), Eucerotoma Laboissiere, 1939 and Neobrotica Jacoby, 1887 – all of which are usually placed in sister-section Cerotomites Chapuis, 1875. Gynandrobrotica has been suggested to be more related to these taxa in some phylogenetic analyses (Eben and Monteros 2004, Gillespie et al. 2008). Other characteristics common to these genera, such as the small eyes, the elongated frons and different shapes of antennomeres I-III should be further studied in order to verify if they are indeed homologues.

There are also three other species with distinctive head features in the male: Isotes onira (Bechyné & Bechyné), 1961 has an enlarged head from the vertex up to the antennal insertions – Figure 3), Diabrotica serroazulensis Bechyné & Bechyné, 1962 has an enlarged labrum, which is as large as half of the frontal length, and Acalymma cornutum (Baly, 1886), has a deep cleft in the frons, with lateral projections, and also a labrum with an acute anterior projection that reaches the frons (illustrated in Munroe and Smith 1980).

Galerucines commonly display filiform antennae, which can show numerous dimorphic variations (Jolivet and Verma 2002, Mohamedsaid 2004). This is also true for Diabroticites, with the main antenna type being filiform and antenommeres mostly subequal in size and shape (Figure 4A).

Non-dimorphic modifications are often related to the length of some antennomeres and sometimes support generic definitions. For instance, the two largest genera in Diabroticites, Diabrotica and Synbrotica are essentially distinguished by the length of antennomere III, described as being subequal to II in the first (same as observed in genus Cochabamba in Figure 4B), and subequal to IV (that is, almost twice as longer as II, as in Figure 4C) in the latter (Bechyné 1956).

Eight genera have diagnostic characters based in their dimorphism of antennae (listed in Table 3). Examples of modified antennae are showed in Figure 4C–F. Because of their variability, dimorphic antennae are the most used structure in descriptions, but their modifications have been scarcely detailed. As a result, there are different genera proposed on characters depicted in sentences such as “apical antennomeres modified” (Cornubrotica), “antennomeres VII and IX of complicated shape” (Buckibrotica – Figure 4E), “antennomeres V-VII modified” (Paratriarius).

The absence of unified criteria in the understanding of what a “modified” antennomere is has lead, several times, to the establishment of artificial grouping of species, simply because a single “aberrant” antennomere can display an assembly of four different aspects of its morphology. Modifications include change in (in quoting marks, expressions used on original descriptions): length – antennomeres considered “elongated” or “shortened” when compared to the usually fixed antennomeres I and III; width – antennomeres described as “swollen”, “inflated” (homogeneous modification), “distally expanded” (heterogeneous modification), and dorsoventrally “flattened”; shape – those described as having “lateral projections” or “ventral excavations”; structure – antennomeres with “rough punctuation” and different amounts of hairs/sensillae. As variations observed in the species do not always correspond to the diagnosis defined for genera they have been included in, these characteristics do not provide an accurate guide to the identification of taxa in Diabroticites.

For the non-monotypical genera based on antennal dimorphic features, most original definitions do not correspond to their actual characters. In Anisobrotica, for example, the “widened” apical antennomeres do not always appear – Anisobrotica binisculpta Bechyné & Bechyné, 1969 only has in common with the other taxa the excavation present in glabrous ventral surface of antennomeres IX-XI (such as excavations observed in apical antennomeres of Anisobrotica donckieri (Baly, 1889) in Figure 5). The same happens with Paratriarius, which includes several species that do not show “modified antennomeres V-VII” present in type-species Paratriarius dorsata (Say, 1824) (illustrated in Wilcox 1965) such as Paratriarius batesi (Baly, 1859), Paratriarius falvolimbata (Erichson, 1847), Paratriarius verrucosa (Jacoby, 1880), Paratriarius alternans (Weise, 1916), Paratriarius nigrotibialis (Bowditch, 1911), Paratriarius castanea (Bowditch, 1911), and also other four species studied by Mohamedsaid and Furth (2011). Instead, these taxa show antennae very similar to the general pattern seen in Diabrotica. The two species included in Cornubrotica do not show identical antennomeres VIII and IX, although both always have ventral excavations (illustrated in Bechyné and Bechyné 1969 and Moura 2005). In Paranapiacaba, the antennal character chosen was, unfortunately, a general one: male antennomeres III-XI uniformly “thickened” (in contrast with slightly slender antennae of females). Nevertheless, antennae do seem to vary uniformly in one genus. In Aristobrotica, the pattern of antennomeres III-V “thickened” is constantly repeated, followed by an unmentioned presence of larger punctuation (Figure 4D).

Although Maulik (1936) suggested that, for indo-asian galerucines, the basal antennomeres are more frequently the altered ones, a result that has been corroborated by Mohamedsaid (2004), that feature does not apply to diabroticites analyzed. Also, no obvious topological pattern is seen in the variation of antennomeres (Table 4).

The number of modified antennomeres oscillated between 1 to 6. Antennomere II was recorded as dimorphic only in species of Aristobrotica, such as Aristobrotica angulicollis (Erichson, 1878) (Figure 4D) and in Isotes onira (Figure 4C). Few modifications occur in antennomeres II and XI. The most affected are antennomeres V to IX. However, there is no indication of an explicit dependency of occurrence between any pair of modified antennomeres. This is the opposite of what has been observed for asian Galerucinae species (Mohamedsaid 2004). Although no pattern is observed, some variation can occur in blocks, i. e., one modified antennomere occurs with one or two adjacent antennomeres also modified.

It seems that most, if not all, antennal variations could be regarded as the result of the presence of punctuation and setae in greater number, either for the production and/or reception of chemical compounds (i. e., pheromones) (Jolivet 2007). A study on Diabrotica virgifera Leconte, 1858, for instance, showed that male antennae have a much greater number of sensilla than females, and numerous glandular points linked to the production of chemical compounds allegedly to be attractive to females (Newman Jr et al. 1993). Unfortunately, there is little knowledge on the biology or even on the anatomy of Diabroticites to support this view as a more generalized tendency.

Although a great diversity of dimorphic characters occur in the thorax of several galerucines (Mohamedsaid and Furth 2011), features reported for Diabroticites are limited to elytra, and legs. In elytra, variation occurs only on the apical fifth, and are either depressions or callosities, both which can co-occur with punctuation (as in some species of Isotes – Figure 6, and in Pseudodiabrotica – the only genus supported on an elytral diagnostic character). Many species in genus Paratriarius show elytral dimorphic characters (such as the callosities present in Paratriarius batesi), and, although such features were never used to originally describe it, they have been used to support the definition of genus Chanchamayia Bechyné, 1956, now considered to be a subgenus in Paratriarius (Smith and Lawrence 1967, Wilcox 1972).

Only metathoracic legs lack alterations in male diabroticites. In accordance with the more general pattern, morphological differences in the pro- and mesothoracic legs usually are connected to the augmentation of absolute size in femora and tarsomeres I, the latter which are directly linked to the partial or total covering of the ventral surface by adhesive setae (Figure 7).

Tibiae and femora can also be modified, being greatly enlarged (such as in Zischkaita serrana Moura, 2003 – Figure 8), and frequently with internal margins concave or bearing tubercles, forming the “prehensile organ” (Bechyné 1956). Bechyné’s concept of such structure is based on a combination of multiple adaptations and should be used with caution, since the homology of the “prehensile organs” can be difficult to assess. Aristobrotica, for instance, has been described as with one diagnostic feature: the “special build of the median tibiae in male”. The detailed analysis of species included, however, indications that there are at least two distinct types of “prehensile organs” being treated as the same modification. While type-species Aristobrotica angulicollis (Figure 9) bears only a concave mesotibiae with laterally flattened apex, Aristobrotica mirapeua Moura, 1997 and Aristobrotica capillosa Moura, 2011 (both illustrated in their original descriptions) display small projections of the ventral margin of the mesofemora (apical in Aristobrotica mirapeua and basal in Aristobrotica capillosa), and differently shaped tibiae (with a basal concavity in Aristobrotica mirapeua and slightly concave tibiae with apical flattening in Aristobrotica capillosa).

Moreover, general characters have been used to support definition of genera such as Cornubrotica and Synbrotica, a genus which is now a synonym of Isotes. The former was supposed to be distinguished by pro- and mesothoracic legs of males without emargination, which is rather common in the tribe, and the latter is characterized by antennomere III elongated and a “uniform pilosity covering the ventral surface of tarsomeres in both sexes” (freely translated from the original, in German) – something that does not accurately identify the males in this group, as they normally have distinctive adhesive disks in their tarsomeres.

The most common abdominal modification seen in some galerucines is the presence of processes with different shapes. Although no abdominal characters aid the definition of diabroticites genera, one character was observed for a single species in the group: a central triangular projection, postero-ventrally oriented, in the posterior margin of the ventrite I, in Zischkaita serrana (Figure 10). A similar alteration is observable in Hemygascelis longicollis Jacoby, 1896, an asian species that belongs to section Phyllobroticites in subtribe Luperina, a group thought to be a sister group of Diabroticina (illustrated in Mohamedsaid and Furth 2011).