It has long been known that termites (Isoptera), cockroaches (Blattaria), and mantids (Mantodea) are closely related (Wheeler, 1904; Walker, 1922; Marks and Lawson, 1962); they are commonly grouped as suborders of the order Dictyoptera (Kristensen, 1991). Although there is a general agreement on the monophyly of the order, during the past two decades the sister group relationships of these three taxa and the position of woodfeeding cockroaches in the family Cryptocercidae in relation to termites have been lively points of debate (see Nalepa and Bandi, 2000; Deitz et al., 2003; Lo, 2003 for further discussion). A variety of factors contribute to obscuring the relationships. First, fossil and molecular evidence indicate that these taxa radiated within a short span of time (Lo et al. 2000; Nalepa and Bandi, 2000). A rapid proliferation and divergence of the early forms would obscure branching events via short internal branches separating clades, instability of branching order, and low bootstrap values of the corresponding nodes (Philippe and Adoutte, 1996;Moore and Willmer, 1997). Second, heterochrony played a major role in the genesis and subsequent evolution of the termite lineage (Nalepa and Bandi, 2000). It is notoriously difficult to determine the phylogenetic relationships of organisms with a large number of paedomorphic characters (Kluge, 1985; Rieppel, 1990,1993). Reductions and losses make for few morphological characters on which to base cladistic analysis, and parallel losses of characters by developmental truncation make it difficult to distinguish between paedomorphic and plesiomorphic traits (discussed in Chapter 2). Third, cockroaches in the particularly contentious family Cryptocercidae live and die within logs and have left no fossil record. Fourth, extant lineages of Dictyoptera represent the terminal branches of a once luxuriant tree, with many extinct taxa. Finally, several phylogenetic studies of the Dictyoptera have been problematic because of ambiguous character polarity, inadequate taxon sampling, and questionable reliability of the characters used for phylogenetic inference (for discussion, see Lo et al. 2000; Deitz et al., 2003; Klass and Meier, 2006).
The bulk of current evidence supports the classic view (Cleveland et al., 1934; Grasse
Fig. 9.1 Phylogenetic tree of Dictyoptera, after Deitz et al. (2003). Mantids branched first, Blattaria is paraphyletic with respect to the examined Isoptera (Mastotermitidae, Kaloter-mitidae, Termopsidae), and Cryptocercidae is the sister group to termites. The study was conducted utilizing the same morphological and biological data base used by Thorne and Carpenter (1992), however, polarity assumptions and uninformative characters were eliminated, characters, character states, and scorings were revised, and seven additional characters were added. The tree suggests a single acquisition of both symbiotic fat body bacteroids (Blattabacterium) and hindgut flagellates within the Dictyoptera. Bacteroids were subsequently lost in all termites but Mastotermes; oxymonadid and hypermastigid flagellates were lost in the "higher" termites (Termitidae—not included in tree). The sister group relationship of Cryptocercus and Mastotermes is supported by phylogenetic analysis of fat body endosymbionts (Fig. 5.7) and the cladistic analysis of Klass and Meier (Fig. P.1). *Blattaria denotes Blattaria except Cryptocercidae.
and Noirot, 1959) that Cryptocercidae is sister group to termites. It is not, however, a basal cockroach group as proposed by most early workers (e.g., McKittrick 1964, Fig. 1). Mantids branched first, with Cryptocercus + Isoptera forming a monophyletic group deeply nested within the paraphyletic cockroach clade (Fig. 9.1; see also Fig. P.1 in the Preface and Fig. 5.7). These relationships are supported by morphological analysis (Klass, 1995), by analysis of morphological and biological characters (Deitz et al., 2003; Klass and Meier, 2006), by Lo et al.'s (2000) analysis of three genes, and by Lo et al.'s (2003a) analysis of four genes in 17 taxa, the most comprehensive molecular study to date. The fossil record and the clocklike behavior of 16S rDNA of fat body endosymbionts in those lineages possessing them indicate that the radiation of mantids, termites, and modern cockroaches (i.e., without ovipositors) occurred during the late Jurassic-early Cretaceous (Vrsansky, 2002; Lo et al., 2003a).
This phylogenetic hypothesis provides a parsimonious explanation for several key characters of Dictyoptera. An obligate relationship with Oxymonadida and Hyper-mastigida flagellates in the hindgut paunch first occurred in an ancestor common to Cryptocercus and termites, and was correlated with subsociality and proctodeal trophal-laxis (Nalepa et al., 2001a). These gut flagellates were subsequently lost in the more derived Isoptera (Termitidae). Endosymbiotic bacteroids (Blattabacterium) in the fat body were acquired by a Blattarian ancestor, or acquired earlier in the dictyopteran lineage and subsequently lost in mantids. All termites but Mastotermes subsequently lost their Blattabacterium endosymbionts (Bandi and Sacchi, 2000, discussed below). The phylogenetic hypothesis depicted in Fig. 9.1, then, is consistent with a single acquisition and a single loss of each of the two categories of symbiotic associations. Eusociality evolved once, from a subsocial, Cryptocercus-like ancestor.
Lo (2003) offers two reasons for exercising some caution in the full acceptance of this phylogenetic hypothesis. First, for two of the genes that support the sister group relationship of Cryptocercus and termites, sequences are unavailable in mantids because they possess neither: 16S rDNA of bacteroids and those coding for endogenous cel-lulase. Second, because cockroach classification is in flux and taxon sampling is still relatively poor, additional data may alter tree topology. One possibility is that mantids may be the sister group of another lineage of cockroaches, which would render modern cockroaches polyphyletic with respect to both termites and mantids (Lo, 2003). Based on their examination of fossil evidence, Vrsansky et al. (2002) suggested that contemporary cockroaches may be paraphyletic with respect to Mantodea as well as Isoptera.
The ancestor common to all three dictyopteran taxa was almost certainly cockroach-like (Nalepa and Bandi, 2000). Cockroaches are the most generalized of the or-thopteroid insects (Tillyard, 1919), while Mantodea are distinguished by apomorphic characters associated with their specialized predatory existence. Both cockroaches and termites have predatory elements in them, although in termites it is probably limited to conspecifics (i.e., cannibalism). Mantids have short, straight alimentary canals (Ramsay, 1990), and like other predators (Moir, 1994), they neither have nor require gut symbionts. Elements of certain mantid behaviors are evident among extant cockroaches, such as the ability to grasp food with the forelegs (Fig. 9.2), and in some species, assumption of the "mantis posture" during intraspecific fights. A cockroach combatant may elevate the front portion of the body, raise the tegmina to 60 degrees or more above its back, fan the wings, and lash out with the mandibles and prothoracic legs (WJB, pers. obs.). Mantids, however, tend to lead open-air lives (Roy, 1999), and although some are known to guard egg cases, the suborder as a whole is solitary (Edmunds and Brunner, 1999).All extant termites, on the other hand, live in eusocial colonies, and have highly derived characters related to that lifestyle. There is little
doubt that the evolution of eusociality was the event that rocketed the termite lineage into a new adaptive zone. A correlate of universal and complex social behavior among extant termites, however, is the difficulty in developing models of ancestral stages based on characters of living Isoptera. Because the best-supported phylogenetic hypotheses have termites nested within the Blattaria, we have license to turn to extant cockroaches, and in particular to Cryptocercus, in our search for a phylogenetic framework within which termite eusociality, and thus the lineage, evolved. It is a big topic, and one that can be explored from several points of view. Here we take a broad approach. We first examine how a variety of behaviors key to termite sociality and colony integration have their roots in behaviors displayed by living cockroach species. We then focus on cockroach development, its control, and how it can supply the raw material for the extraordinary developmental plasticity currently exhibited by the Isoptera. We address evolutionary shifts in developmental timing (heterochrony), and how these played crucial roles in the genesis and evolution of the termite lineage from Blattarian ancestors. We then turn to proximate causes of termite eusociality, first discussing how a wood diet and the symbionts involved in its digestion and as similation provide a framework for the social transition. Finally, using young colonies of Cryptocercus as a model of the ancestral state, we show how a simple behavioral change, the assumption of brood care duties by the oldest offspring in the family, can account for all of the initial, defining characteristics of eusociality in termites.
Was this article helpful?
You Might Start Missing Your Termites After Kickin'em Out. After All, They Have Been Your Roommates For Quite A While. Enraged With How The Termites Have Eaten Up Your Antique Furniture? Can't Wait To Have Them Exterminated Completely From The Face Of The Earth? Fret Not. We Will Tell You How To Get Rid Of Them From Your House At Least. If Not From The Face The Earth.