James M. Carpenter

Choosing Among Multiple Equally Parsimonious Cladograms

With the realization that multiple equally parsimonious cladograms might exist for a given data set (Mickevich, 1978), construction of classifications in such instances became a problem. Consensus trees (Adams, 1972; Nelson, 1979), originally developed for producing a “compromise classification” (Adams, 1972) between cladograms produced from dzferenl data sets, is one possible solution. The consensus tree, representing the information on grouping shared by all the competing cladograms, might be viewed as a “conservative” classification. Such a viewpoint has been advocated by proponents ofdistance techniques in the case ofmultiple trees ofnear-optimal fit (e.g. Prager and Wilson, 1978). In this case, such a procedure is a misapplication. The consensus tree, being less resolved than any of the cladograms from which it is calculated, has less explanatory power (Farris, 1983) than any of them, and so any of the competing cladograms would br a better choice as a phylogenetic hypothesis/classification. Mickevich and Farris (1981) and Miyamoto (1985) make similar points. The question is then how to choose among the cladograms, if this can be done. Recently, Brooks et al. (1986) have addressed the issue of choice among multiple equally parsimonious cladograms. However, their discussion is deficient. They considered three statistics: the consistency index (Kluge and Farris, 1959), F-ratio (a simple function of Farris’ (1972) f statistic) and the D measure, an application of the Shannon entropy statistic from Gatlin (1972). The consistency index does measure evidential support for phylogenetic hypotheses, but this is not true for the other measures. The f statistic, developed for application to distance analyses, measures the pairwise homoplasies for a given tree, and, to quote Farris (1983:22), “the pairwise homoplasies are not independent”. This measure is thus not closely related to the concept of evidential support underlying phylogenetic analysis (Farris, 1983), and, in fact, a tree optimizing f (hence the F-ratio) may not be most parsimonious for a given data set. Brooks et al. (1986) make that last observation, but do not then provide any rationale for using the F-ratio. Similar comments apply to their use of D, which they characterize as measuring the “ i n z a t i o n content of the constraints” (Brooks et al., 1986:572). They provide an example (their fig. 5) where D is optimal for a nonparsimonious cladogram, which is reason enough to reject this measure. Brooks et al. argue that D can distinguish between autapomorphies and synapomorphies, whereas the consistency index cannot. Their discussion of the consistency index is misled. Whereas this index is inflated by inclusion of invariant and autapomorphic characters, plainly it should be calculated with such features excluded in a proper cladistic analysis-a point pheneticists have made (Colless, 1983). Thus, the example shown in their fig. 6a-c has an identical consistency index for all three cladograms calculated over all the characters. But calculated only on informative characters, their cladogram 6b

TESTING SCENARIOS: WASP SOCIAL BEHAVIOR

Abstract— A complex evolutionary model is tested with a cladistic approach. Cladograms constructed for all of the genera of social Vespidae are optimized for characters associated with social behavior. The character state assignments to the interior nodes are compared with the stages envisioned in the “polygynous family” hypothesis for the evolution of wasp social behavior (West‐Eberhard, 1978). Several of the transitions proposed in the model are consistent with the results: caste formation preceding long‐term monogyny, and long‐term polygyny developing from monogyny. Some transitions do not accord with these results: long‐term polygyny evidently did not evolve directly from a “rudimentary‐caste‐containing” stage, and a stage of tasteless nest sharing may not have occurred.

Phylogeny of the Dictyoptera

Abstract. Relationships among six key dictyopteran taxa (Mantodea; Blattodea (excluding Cryptocercidae); Cryptocercidae; Mastotermes darwiniensis, Termopsidae and Kalotermitidae [Isoptera]) are analysed based on seventy morphological, developmental and behavioural characters. The fossil record and the ‘living fossil’ genera Cryptocercus, Mastotermes and Archotermopsis are discussed in detail. Exact analysis of the character state matrix by implicit enumeration (Hennig86) resulted in one cladogram, strongly supporting Blattodea + Cryptocercidae as a sister group to Mantodea, with the Isoptera as a sister group to that complex. Arrangements within the termites are equivocal, with Termopsidae and Mastotermes darwiniensis possible as the relatively most primitive element of Isoptera.

Phylogenetic relationships and classification of the Vespinae (Hymenoptera: Vespidae)

Abstract. The phylogenetic relationships of the genera, subgenera and species‐groups of the Vespinae are analysed using cladistic techniques. The results are used as the basis for a natural classification of these wasps. The cladogram for the four genera recognized is: Vespa+(Provespa + (Dolichovespula + Vespula)). No subgenera are recognized; all those previously described are synonymized with the appropriate genus. The synonymies of Nyctovespa with Vespa and Rugovespula with Vespula are new.

Phylogenetic analysis and the evolution of queen number in eusocial Hymenoptera

Analyses of the evolution of colony queen number in eusocial insects have generally been conducted without specific reference to phylogenetic relationships, leading to incomplete evolutionary explanations for this key attribute of social organization. Consideration of queen number in a phylogenetic context in the highly eusocial Hymenoptera reveals that its evolution has been very conservative in the bees but that it is a highly labile character in most ants. The wasps appear intermediate in this respect, with some large and widespread clades characterized by little or no phylogenetic variability in queen number. This hierarchy of phylogenetic lability suggests that while ant populations may often be responsive to selection on colony queen number linked with local ecology, bees and wasps appear less responsive in this regard, with a significant element of phylogenetic conservatism involved in the expression of this social trait in the latter two groups.

A character analysis of the North American potter wasps (Hymenoptera: Vespidae; Eumeninae)

A cladistic analysis of the nearctic genera of Eumeninae is presented. The ground plan states of 43 character systems are discussed, and the first comprehensive cladogram for any significant portion of the subfamily is presented. At least eight of the 26 genera are apparently nonmonophyletic: Zethus, Montezumia, Euodynerus, Odynerus, Pterocheilus, Stenodynerus, Leptochilus and Microdynerus. A key to the nearctic genera accompanies the text.

A synonymic generic checklist of the Eumeninae (Hymenoptera: Vespidae)

The present work is an extension of a similar list in Carpenter (1983), and arose from preparatory work for a phylogenetic analysis of nearctic potter wasp genera (Carpenter and Cumming, 1985). The most recent available world list of genera is over 80 years old (Dalla Torre, 1904), and fully 57% of the genus-group names currently used in the Eumeninae have been proposed since Bluethgen (1938; for more detail on the history of eumenine taxonomy see Carpenter and Cumming, 1985). The following checklist includes all the currently recognized genera of Eumeninae sensu Carpenter (1981), with their synonyms and subgenera. The arrangement is alphabetical based upon most recent usage, and incorporates the decisions pertaining to eumenine generic nomenclature rendered by the International Commission on Zoological Nomenclature (ICZN) in Opinions 747 (1965), 893 (1970) and 1363 (1985). The format is basically that of Krombein et al. (1979). The original citations are followed by the type species designation. Synonyms, and subgenera with their citations and synonyms, are listed after this; the nominotypical subgenera are not listed separately. Where two dates are listed, the first is the true date of publication, whereas the date listed in parentheses is that printed on the paper. A misspelling is indicated by (!), and quotation marks are used for incorrect names. No effort has been made to list all misspellings; only those occurring in works considered important. Nomenclatural changes derive from ongoing work on a catalog of neotropical eumeninae (with J. van der Vecht) and a generic reclassification of this group Neodiscoelius Stange is a junior objective synonym of Protodiscoelius Dalla Torre (new synonymy); Cephalastor Soika is raised to genus (new status), and its type species, Hypalastoroides depressus Soika, synonymized with Odynerus relativus Fox. In addition, type species are designated for

CLADISTICS OF CLADISTS

Mishler (1987:58) stated that he could not define cladistics on the basis of a set of shared ideas, dismissing synapomorphy and monophyly because the evolutionary connotations of this traditional Hennigian idea have been denied by Patterson and others and various workers outside the Hennigian school have adopted one or more of Hennigs concepts. As if particular evolutionary connotations were necessary for the evidential use of these concepts, or acquisition of one (not both) by other workers somehow vitiated the defining power of their joint possession. But a similar stance underlies the distinction between pattern and phylogenetic cladistics. Hull (1979, in a published version of a 1977talk) first distinguished these types of cladistics, when he attributed to Nelson the notions of atemporal, big C cladistics and historical, little c cladistics. Interestingly, in a talk he gave at Cornell in 1982, Hull had a separate category for quantitative cladistics which was orthogonal to the other two; apparently the pattern/ phylogenetic dichotomy was insufficient to classify cladists. But of course Beatty (1982) was the one who really made the distinction notorious. He basically followed Hull in splitting cladists into pattern and phylogenetic, but added the accusation that the former were antievolutionary. Many authors besides these have since adopted the pattern versus phylogenetic distinction, although those stated to be pattern cladists have universally denied that there is any methodological difference. No one has ever shown any case where different results are obtained if you are one of those antievolutionary pattern cladists or one of Darwins (or Prigogines) apostles. But now its being claimed that pattern cladism limits the success of cladistics. Brooks and Wiley (1985) concluded that the dichotomy was a matter of extending cladistics to probe causality or not. This is the veriest nonsense-as if pattern cladists cant interpret their results as evolutionary, which may consist of no more than hanging ornaments on their trees. Distinctions based on intent have been tried several times in discussions of systematic schools, but the results have not been consistent. A review of the cladistics of cladists illustrates this. First, each possible resolution of the usual three schools has been proposed. Hull (1970) proposed: pheneticists + (syncretists + cladists); the latter taxa were grouped by the use of phylogeny in classification. Dupuis (1984) suggested: syncretists + (pheneticists + cladists); the latter were grouped because they work on the products of evolution whereas syncretists work on processes (and so are typologists!). He explicitly denied any difference in principle between Nelsonians and Hennigians, by the bye. The third resolution of this three taxon statement is based on practice rather than intent. Mayr (1974) and Platnick (1985b) both proposed: cladists + (pheneticists + syncretists); the latter two classify by a phenetic approach. Here we have an arch syncretist and an archcladist agreeing on S01TI£thing. It TTU/St be true.

The Phylogenetic System of the Gayellini (Hymenoptera: Vespidae; Masarinae)

The Gayellini is one of the two. tribes of Masarinae (Carpenter, 1981). Endemic to the Neotropics, the majority of the species are Patagonian, but one ranges as far north as Mexico. With ten described species, the group is far less speciose than its sister-tribe Masarini, which has over 200 described species (cf. Richards, 1962), and most species are rarely collected. These wasps have a very distinctive appearance among Vespidae (Fig. 1), and their taxonomic history has been more turbulent than any other higher vespid taxon. Although the phylogenetic placement of the group as a whole has now evidently been settled (Carpenter, 1981), no study has been made of the species. The current generic classification is fragmented, and there have been no keys to all of the taxa. In this paper, I investigate the phylogenetic relationships of the species, and present a revised generic classification along with keys to all taxa.

A review of the subspecies concept in the eumenine genus Zeta. (Hymenoptera: Vespidae)

Menke and Stange (1986) reported the establishment of the potter wasp Zeta argillaceum argillaceum (L.) in Dade County, Florida. This is the first species of this neotropical genus to be found in the United States. While identifying a small collection of Florida eumenines for Dr. Peter J. Landolt of the USDA Insect Attractants Lab in Gainesville, I discovered an additional specimen of this species from Dade County. It represents a color form, or subspecies, different from that reported by Menke and Stange (1986). In this note I discuss recognition of subspecies in Zeta, and demonstrate that most are artificial taxa. Plasticity in coloration is a common phenomenon in Hymenoptera, and has even been shown to be experimentally manipulable in Vespidae (MacLean et al., 1978). Yet recognition of subspecies purely on the basis of color differences remains a common practice in vespids, including Zeta, and other aculeates. The recent revision of the genus by Giordani Soika (1975) recognizes only four species in Zeta, but these are divided into no fewer than 15 subspecies, 10 of which are in argillaceum? These are all based on color, and in the continental forms the distributions frequently overlap. I have examined most of these subspecies, and conclude that they do not merit formal recognition. Most of them are synonymized below. I have not treated the subspecies of the Antillean abdominale. Acronyms for collections are those of Heppner and Lamas (1982).