Who Said Cockroaches Are Not Colourful!

On a recent trip with some James Cook University students, a few remarked that they had no idea tht cockroaches could be so colourful.

Well here are a few examples that will hopefully challenge your perception of these insects as being rather dull creatures that often invade human habitation. The fact is that most species have little to do with humanity and could not survive long in our homes, restaurants, hotels etc. We just don't
"offer" what they are looking for!

This one, the Inland Ellipsidion, Ellipsidion amplum Hebard literally flew into me while I was walking on a Cairns street in the centre of town. It is a diurnal species that lives on flowers where it feeds on pollen and nectar. The bright orange colour pattern suggests Mullerian Mimicry.

The Inland Ellipsidion Ellipsidion amplum Hebard, dorsal and ventral views.

Ellipsidon cockroaches are fairly common across the continent where they have a rather similar appearance. This is among the largest of species measuring 17 mm.

The above-noted specimen was a bit battered after being carried around in a shopping bag for a few hours. Here is an example of the species in much better condition. Note the orange-tipped antennae. Other species have the antennae with white tips. In addition, the antennae are very hairy (click on the photo to enlarge), a condition quite unusual in most Australian cockroaches.

Ellipsidion is a member of the large family Ectobiidae (formerly Blattellidae) and is in the subfamily Pseudophyllodromiinae. This is quite a mouthful. However, species in this subfamily are usually very attractive. Here are a few examples.

 The Australian Allacta, Allacta australiensis Roth. 

This species lives in rainforests where it can be found only on tree trunks. It seems to prefer certain trees and not others. It measures 11 mm and is active at night.

 The Variable False Pseudobalta, Pseudobalta cinctella (Hebard)

Pseudobaltas are tropical cockroaches that are active at night. They have been found under bark during the day. The species has a broad distribution across the continent and the markings on the thorax (pronotum) are variable. It is small, measuring about 6 mm in length.

 The genus Balta includes some 39 described species in Australia. There are many undescribed species, like the one above. Baltas are distinguished on the basis of colour and morphological structures. They are nocturnal and live in leaf litter during the day. Their numbers are often great and they probably serve as food for many other organisms. All known species are active after dark.

 The Pretty Balta, Balta verticalis Hebard.

This Balta is aptly named. The specific name refers to the stripes between the eyes. It is a fairly common tropical species in northeast Australia where it can be found in rainforests and along rainforest margins. It measures about 12 mm in length.

 The Grey Balta, Balta scripta (Shelford)

One of the commonest Baltas in the northern Tropics, this species spends the day in leaf litter and is active at night. It frequently comes to lights. The pattern on the pronotum and greyish colour is very characteristic. 

The Small Mediastinia, Mediastinia australica (Shelford)

Mediastinias are more common than previously thought. They are very small, measuring less than 5mm in total body length. They are very colourful and have a very flattened appearance. This facilitates diurnal life in leaf axils and unfurling leaves. After dark they emerge to feed on particulate matter found on leaf surfaces.

The variable Speedy Cockroach, Choristima bimaculata Roth

This cockroach is another nocturnal species but it  belongs to another ectobiid subfamily, the Ectobiinae. Seldom seen during the day, adults can be attracted to light late in the evening. This species has a number of different morphs based on the pattern of the pronotum and wings. They are small, measuring less than 5 mm in total length.

Sloane's Northern Wingless Cockroach, Cosmozosteria sloanei Shaw.

This is an example of the most characteristic group of Australian Cockroaches, the subfamily Polyzosteriinae of the Blattidae. Everyone who has wandered around in the outback or the mixed woodlands of the tropics has come across these large cockroaches abroad during the day. They are very conspicuous in their colour patterns and they seem to be walking bags of nasty chemicals. They are avoided by birds and lizards. This one is about the size of a 20 cent Australian coin.

But these insects are the topic of a future blog.

The Complexities of Cockroach Systematics

Australia must be the centre of diversity for the cockroaches. It seems more species occur here than any other known place. More than 500 species have been described and it is estimated that half again that number await discovery and description. Just on this 1.6A site of regenerated rainforest here in Kuranda, I have recorded more than 90 species.

This bar graph shows the distribution of families against the number of species discovered. The Ectobiidae (formerly the Blattellidae) have more than seven times the next family (Blaberidae) represented.Within the Ectobiidae, there is a a rather balanced representation of subfamilies known for Australia.

The purpose of this blog is to relate some of the complex problems that a taxonomist faces when working in this ancient group of insects.

Cockroaches have been around for a long time. Fossil evidence suggests that cockroaches 200 million years ago looked pretty much as they do today. Being generalised feeders rather than specialised ones, they can find what they need most of the time. As a result they have occupied many and varied habitats. Within Australia, they can be found almost anywhere from the mountain tops, deserts and coastal islands. A few seem to be subaquatic.

Most entomologists who study cockroaches do so from collections in the great museums of the world. Some have never seen a cockroach in nature, except for some of the 6-8 species that exist in the kitchens and dwellings of humanity. These few domestic species have given the entire group a bad name. But the thousands of species that occupy natural habitats are interesting and their biologies varied.

Louis M. Roth photo: P. Naskrecki

Lou Roth spent a lifetime studying cockroaches. He probably named more species than anyone else. He had a "worldwide" view of cockroach systematics. That means that he knew whether genera were confined to rather restricted localities or covered vast expanses of territory that might include several continents.

In the end this may have contributed to a dilemma. Dr Roth spent many of his last years working on the Australian fauna. The reasons were simple: Australia has a varied and complex diversity of species and there was money around to help pay for the illustration and publication of his results. [He was active in the 70's and 80's when the Australian Biological Resources Survey was well funded and well-regarded by thoughtful governments concerned about the Australian biota.]

Lou published many generic revisions. His last had to do with cockroaches in the ectobiid subfamily Blattellinae. Several genera are worthy of attention here. They illustrate the complex nature of our fauna and the tough taxonomic decisions that await cockroach-ologists!

Three genera seemed to perplex Lou. They are Carbrunneria, Johnrehnia and Beybienkoa. [These are all named after famous cockroachologists, but that's another story.] Basically the three genera are distinctive in the following way:

"specialised" here means bearing a glandular opening or a patch of hairs.

Carbrunneria; In males first abdominal tergite (the dorsal part of the abdomen) not specialised; seventh tergite specialised;

Beybienkoa; In males the first abdominal tergite is specialised; seventh tergite not specialised;

Johnrehnia; There are no modifications on the tergites of males at all.

In compiling details heading towards the publication of this book,

I discovered some problems that suggest that the above genera need a bit of reorganisation. Perhaps, there are more genera that at first expected. All seem to occur in Australia and a few extend into New Guinea. All are nocturnal and live in leaf litter or under bark by day, and emerge after dark to feed.

At any given locality, all three genera may be encountered and one can frequently find more than one species in a given genus.

Let's examine two species in the genus Johnrehnia, found recently in mixed open forest north of Mareeba, Queensland. These species illustrate the extreme differences in appearance of Johnrehnia species as they are presently understood. Remember their distinguishing feature is that the males have no glandular openings on the dorsal abdominal tergites. Both species are undescribed and both very common.

This dark motif is present on may Australian species. Males can be told at a glance by the tawny colour of the wings.

 Females are uniformly black on top.

With cockroaches, especially the species in the Blattellinae, the male subgenital plate is differentiated and species' distinctive.

 Here we see the tip of the male's abdomen upside down. So his right is to the left. We an see two "styles". These are peg-like sensory devices that are on the margin of the subgenital plate. Taxonomists look at these structures first because they are distinctive and do not vary much form one individual to another.

In many cockroaches, especially the Blattellinae, the male genitalia are not bilaterally symmetrical. This just means that the right side is different from the left. The styles are a good example.

This is the dorsal view of the male showing the tenth tergite (the last body segment) and the subgenital plate with the highly modified, spiny right style and the more or less "normal" left style.

The elongate slender structures are bits and pieces of the concealed genitalia which are located under the subgenital plate and within the end of the abdomen.


Let's look at another Johnrehnia species found at the same locality on the same night.

This species of Johnrehnia, like the one above, has many examples that look just like this-golden colour, distinctive thorax (pronotum) yellow legs and head with a black band across the top and the face with a distinctive number of spots or bars.

Head of above species.

 Dorsal view of the male abdomen. This shows a pair of projections separated by a flexible tongue. Structures can be present or absent and varies in shape and direction depending upon the species. The surface is so colourless that you can see through it to the concealed genitalia within the abdomen.

Here we see the tip of the abdomen on end with the subgenital plate with a greatly modified, hook-like right style and the small left style which is "somewhat" modified.

The male subgenital plate removed. Ventral view showing a bit of the right style and most of the smaller left style.

An end on view of the male subgenital plate. The shape of the hook-like right style is very distinctive of this species. Others have it shorter and with a different base or twisted in a different direction.

What is the purpose, if any, of these outlandish modifications. There is an old "theory" called the "lock and key theory" which suggests that the structures borne by the male can only fit into specifically modified pockets within the abdomen of the female. This is said to maintain species' integrity. With many species of a given genus often present at any given locality, this could act to help prevent hybridisation. Cockroaches probably use chemicals to initially distinguish one another. The glands on the back of other genera probably produce substances that are very species distinctive.

But to conclude this story. It seems that the some of the above very different-looking kinds of Johnrehnia may not be Johnrehnia at all. Some of the species with the "golden appearance" have very faint traces of glandular openings on the seventh tergite. This would point them in the direction of Carbrunneria. 

And it seems that was what Lou Roth was thinking just prior to his death. We have discovered a number of species of the "golden morphs" labelled by Lou as various species of Carbrunneria. Many were previously described species of Johnrehnia. But we could find no written manuscript of Lou's discussing this switch in his thinking and so it will be left to others to sort out this problem.

In the meantime, we continue collecting examples from as many localities as possible to build up the stockpile of evidence so that any changes can be based on an abundance of evidence. DNA sampling can certainly help here.