Insect evolutionary morphology is flooded with axioms that every entomologist is familiar with but perhaps cannot fully understand or visualize. Of course, who has not heard about the mouthparts of entognathous insects, that are retracted into the head capsule or about the male genitalia of Zygentoma, that is really “simplified”.
Illustrations from older papers have been cited and redrawn over and over again for contemporary textbooks and reused in insect morphology lectures. Does this mean that old drawings are so perfect that no additional illustration is needed? Well, they certainly are good enough to deliver simple messages, but are insect morphology axioms really so simple?
During our “Project Snodgrass” we try to visualize anatomical structures using more advanced technologies and then test if they could serve better as educational tools (Andy’s upcoming ENT 432 class will be a good opportunity for that!). We also wonder about the axiomatic nature of these facts—e.g., are entognathous mouthparts really retracted to the head capsule? Is the firebrat male genitalia really so simple?
Lets start with basal hexapods. Most people, perhaps, consider collembolans, proturans and diplurans to be, let’s say, not the most exciting creatures. These soil-dwelling invertebrates have superficially boring morphology: they are small and reduced, and, on the top of this, most of them are white. But these properties actually make them extremely interesting for somebody in possession of a confocal laser scanning microscope (CLSM). Because their cuticle lacks any melanin their white muscle is visible externally, and because they are really small, the laser can penetrate deeply into their bodies. We can easily visualize internal structures.
Non-insect hexapods share some plesiomorphic (ancestral or primitive) morphological traits that can be found in non-hexapod pancrustaceans (“Crustacea”) but never in true insects (Insecta). Perhaps the most well known is the structure of their antenna, which is composed of true appendage segments (see Smartbox below).
|Box 1. Appendage segments vs. annuli Insect appendages are encircled and defined by evaginations of the integument (epidermal cell layer and cuticle) that are attached to the body via muscles. The cuticle of most appendages (e.g. antennae and legs) are composed of repetitive, ring-like sclerites (harder plates) that are connected to each other with conjunctiva (more flexible, membrane-like regions). If a ringlike sclerite receives the site of origin of at least one muscle (i.e. musculated), then the appendage portion defined by the sclerite is called appendage segment, if the sclerite is not musculated, the portion is called annulus (pl. annuli) or mere (meres). In true insects, only the first two ring-like sclerites of the antenna, the scape and pedicel, are segments. The rest of the antenna (the flagellum) is composed of annuli that are often referred as flagellomeres. This antenna type is the annulated antenna. Besides Insecta, numerous crustaceans (e.g. Malacostraca) have annulated antennae (the first malacostracan antenna is composed of two or three basal, musculated appendage segments and a distal, sometimes paired flagellum). Segmental properties of appendage portions are, however, not always fully reflected by the morphological terminology. Ring-like sclerites of mouthparts appendages (maxillary and labial palpi) are usually called palpomeres. Although the “mere” suffix refers non-musculated units, most basal palpomeres are musculated (Fig. Box). Should we somehow differentiate them from the non-musculated palpomeres? A possible solution is provided in a recently published paper from our lab group.
Figure Box. CLSM volume rendered image showing the maxilla of Orthogonalys pulchella, posteromedial view, distal to the top-left, doi: 10.6084/m9.figshare.956281. Image by István Mikó (CC BY 2.0).
Imms (1939) was the first to observe (and publish) that some basal hexapods (diplurans and collembolans) have truly segmented antennae, and he observed the positive correlation between the presence of Johnston’s organ (a mechanosensory organ of the pedicel that detects motion of the flagellum) in annulated antennae and its absence in segmented antennae (Imms 1940).
Imms (1939) illustrated the antenna of numerous arthropod taxa, including Diplura. His illustration has been widely used in the textbooks over the last 70 years (e.g., Chapman 2010). One of the first goals of “Project Snodgrass” was to image and annotate the antenna of a Campodea specimen (Diplura: Campodeidae).
Figure 1. CLSM volume rendered media file showing the antenna of Campodea sp. (Diplura: Campodeidae). Image by István Mikó (CC BY 2.0)
As in many other basal hexapods, the cuticle of campodeids is transparent, so we were able to visualize all but one antennal muscle with the CLSM (Strangely, the same muscle, the depressor of the antenna was the only muscle that was not illustrated by Imms).
Figure 2. CLSM volume rendered media file of the proximal part of the antenna of Campodea sp. (Diplura: Campodeidae). Image by István Mikó (CC BY 2.0).
The campodeid antenna is externally more simplified than most insect antennae. Only trichoid setae are present (their base is connected with ganglionic enlargements that are more advanced in the distal segments with less developed musculature. The first antennal segment (scape) is short and is connected to the head capsule with 4 muscles (internal and extrenal levators, the extensor and the depressor of the antenna (the last one is not illustrated on Fig. 2). It is interesting, that the second antennal segment (pedicel) is also connected to the head capsule (long flexor of segment 3). Segment 3 is moved by only one scapal muscle (flexor of segment 2) because other scapal muscles (basal longitudinal muscle and depressor of segment 3) insert on segment 3. Segment 3, according to Imms (1940) is connected to the labrum by two muscles (short and long extensors of segment 3). Except two diagonal muscles (depressor of segment 4 and 5) the presence of three muscles (dorsal longitudinal and dorsal and ventral extensor) is shared by all segments from segment 3. We did not find the ventral longitudinal muscle, which was described (but not illustrated by Imms 1940).