Recently we brought together much of the literature linking cockroaches with the transmission of certain organisms that cause disease in man and other vertebrates. In that paper (1957a) we concluded that cockroaches, being potential vectors of pathogenic agents, should not be regarded simply as minor annoyances. Obviously the associations of cockroaches with agents of vertebrate diseases are of more immediate importance than their relations with pathogens of lower animals or with nonpathogens. On the other hand, cockroaches are of general economic as well as medical importance, and their control is sought by many who are unaware of their medical significance. That the control of domiciliary cockroaches is far from satisfactory may be inferred from current entomological and pest-control journals in which new insecticides are continually advocated to replace others found to be inadequate. Possibly new approaches to the control of cockroaches are needed. Whether these lie in the direction of increased use of parasites and predators for the biological control of these insects remains to be seen. In any event, the more we know about any insect, especially its ecology, the greater the likelihood of achieving satisfactory control. In order to advance knowledge in any field of science, new research should proceed from the results of prior investigations when these exist. We hope that the observations and experiments cited herein may suggest areas for future research and exploitation.

To the best of our knowledge no previous publication has brought together the vast literature on the parasites, predators, commensals, and other symbiotic associates of the Blattaria. For this reason, we have tried to assemble observations on all such known associations. Undoubtedly we have overlooked some records, as, for example, those buried in papers dealing with other phases of cockroach biology. We hope that such inadvertent omissions will not seriously impair the usefulness of this compilation. Whatever its defects, this review should be a unified source of information for all who are interested in the biotic associates of cockroaches.

In addition to previously published information, this monograph also contains original records and observations on the associations of cockroaches that are reported here for the first time. Although some of the observations were made by us, others were made by colleagues who have graciously made their knowledge available to us in private communications.

HISTORICAL

Chopard (1938) in his book La Biologie des OrthoptÈres reviewed much of the literature on cockroaches, but of the many biotic associations that exist he discussed only the commensal cockroaches, gregariousness, and familial associations. Asano (1937), who reviewed the natural enemies of cockroaches, mentioned about 10 groups of animals that attack cockroaches. Thompson (1951) in his Parasite Host Catalogue, which was based mainly on papers abstracted or noted in the Review of Applied Entomology, listed only 19 insect parasites of cockroaches. Eighteen of these were Hymenoptera which attack only cockroach eggs; the single dipteron listed (Sarcophaga lambens Wiedemann, supposedly parasitic on Pycnoscelus surinamensis) is not a parasite in this case, but deposits its eggs on the dead insects (see p. 229). Cameron (1955) listed as parasites and predators of the cockroach 24 species of hymenopterous egg parasites, 7 species of Ampulex which hunt nymphs and adults, 17 Protozoa, 13 nematodes, 5 bacteria, 2 mites, and a few other miscellaneous predators. In his classified list of the protozoan parasites of the Orthoptera of the world, Semans (1943) listed about 26 species from cockroaches. Linstow (1878, 1889) recorded 14 species of helminths from cockroaches. Van Zwaluwenburg (1928) listed 33 names of roundworms which are commensals or secondary parasites of cockroaches, but some of these names are synonyms. La Rivers (1949) extended this list with 13 additional species. Chitwood (1932) recognized 24 species of nematodes which are primary parasites (probably commensals) of blattids. Steinhaus (1946) gave many instances of biological relationships between cockroaches and bacteria, fungi, and yeasts, but the cockroaches were not discussed as an entity and the information is scattered throughout the book.

In surveying the literature on this subject we have collected a far more extensive list of animals and plants associated with cockroaches than one might have expected from an examination of any one of the previous papers on this subject. In our review of the medically important organisms associated with the Blattaria, we pointed out that in addition to many experimental associations cockroaches have been found to harbor, naturally, 4 strains of poliomyelitis virus, about 40 species of pathogenic bacteria, the eggs of 7 species of pathogenic helminths, and to serve as intermediate hosts of 12 other species of helminths pathogenic for vertebrates; cockroaches have also been found to carry, on occasion, 3 species of Protozoa that are pathogenic to man and 2 species of fungi which are sometimes found associated with pathological conditions.

In addition to the above organisms of medical importance, we have compiled records of other organisms, nonpathogenic to vertebrates, which are naturally associated in some way with cockroaches. None of the following numbers can be considered absolute because some names may be synonyms. However, we believe that these figures are very close to the actual numbers of species that have been isolated because we have attempted to refer all obvious synonyms to the currently accepted name for each organism. On this basis there are about 45 species of bacteria, 40 fungi, 6 yeasts, 90 Protozoa, and 45 helminths that have been found associated naturally with cockroaches. Of the arthropods there are about 2 species of scorpions, 4 spiders, 15 mites, 4 centipedes, and 90 insects. Of vertebrates there are 4 species of fish, 16 amphibians, 12 reptiles, 20 birds, and 27 mammals. Besides these there are many records of experimental associations that have been contrived in the laboratory.

Some idea of the increase in our knowledge of the biotic associations of cockroaches, during the last 70 years, may be gathered from a comparison of the above figures with those of Miall and Denny (1886) who presented " ... a long list of parasites which infest the Cockroach." This list included 2 bacteria, 6 Protozoa (some of the names are synonyms), 7 nematodes (some of these names are also synonyms), 1 mite, 1 wasp, and 1 beetle. In addition, they mentioned as other foes of the cockroach: monkeys, hedgehogs, polecats, cats, rats, birds, chameleons, and frogs.

METHODS

We have listed the organisms known to be associated with cockroaches systematically by phylum, class, order, and family. Within each family the organisms are listed alphabetically by genus and species. Under each organism the associated cockroaches are listed as natural or experimental hosts, vectors, or prey. Identified cockroaches are listed by the currently accepted name. Unidentified cockroaches are indicated by the word "Cockroaches." The name of each cockroach is followed by the country in which the observation was made, the authority for the record, and with a few exceptions[3] pertinent biological information, where this is known. Question marks following the names of organisms or countries indicate tentative or questionable identifications.

Records of predators capturing and feeding on cockroaches in zoos and on shipboard we consider natural, even though it is very likely that these particular predators would not normally have access to this prey in nature.

Experimental prey are cockroaches that were fed to predators in the laboratory. Although these predators may have little, if any, access to these cockroaches in nature, we have included such records to indicate the relative acceptability of cockroaches as food by a wide variety of animals.

Records of presumed or known cockroach associates that give no information about an associated cockroach are not included in this review, even though certain of these (e.g., species of Ampulex, Evania, Podium) probably prey upon or parasitize cockroaches exclusively.

The validity of a host-parasite or predator-prey record is dependent upon the accuracy and knowledge of the observer. In assembling these records we have had to accept, in most instances, the identifications of species made by the original authors. However, as a result of our studies on the biology of various species of cockroaches, including some work on their hymenopterous parasites, we have questioned certain records in the literature. Other dubious records which have been perpetuated from one publication to the next, but which apparently were not based on fact, have also been questioned or have been clarified with the aid of specialists in particular groups.

Because the records cover a period of many years, the names of many of the organisms as well as the names of some of the cockroaches have been changed. Although it would have been comparatively simple to list the names as they appeared in the original references, this would have resulted in misleading redundancy with the same organism being catalogued under several synonyms. We have attempted to list each organism by its currently accepted name. However, no attempt was made to prepare complete taxonomic synonymies; the only synonyms given are those that identify the organisms by the names used by the authors of the papers cited. The synonyms under which the cockroaches may have been cited originally are listed in section II. The synonyms of associated organisms are listed with each organism. Although authorities for the name changes of the cockroaches are given, these workers are not necessarily those who were initially responsible for the synonymies. Various sections have been checked by specialists in the particular groups. Although we have accepted name changes suggested by these reviewers, we assume full responsibility for the names.

FUTURE WORK

After having examined thousands of references on cockroaches, we are impressed by how little is known about the biology of most species. As a conservative estimate there are 3,500 described species of Blattaria (J. W. H. Rehn, 1951). In our literature survey we found records of biotic associations for about 400 species. Unfortunately, many of these records contain only a sentence or two of biological information. Our detailed knowledge of cockroaches is based on studies of the few domiciliary pests that man attempts to eradicate. Comparable studies of the bionomics of the less-well-known species should add much valuable information to our knowledge of this ancient group.

Our understanding of most predator-prey and parasite-host relationships has barely progressed beyond the taxonomic stage. The total effect of predators and parasites in limiting natural populations of cockroaches remains to be determined. It is still not known how, for example, predatory or parasitic wasps select specific cockroaches from among all other insects. Secretions produced by certain cockroaches (e.g., 2-hexenal by Eurycotis floridana) will ward off certain predators. The identities and biological activities of most cockroach secretions are unknown, but the use of protective chemicals against predators may be widespread among cockroaches. If so, how effective are these repellents in protecting the individual or the species? It is not known whether cockroaches are protected by apparent mimetic resemblances to other arthropods. There is no experimental proof that insect parasites can successfully attack the eggs of cockroaches that incubate their eggs while they are being carried by the female.

It is conceivable that biological control of cockroaches might be achieved in limited areas such as man-made structures or sewers, but this possibility has not been thoroughly explored. It would be informative to know what effects, if any, organisms such as bacteria, Protozoa (e.g., gregarines), intestinal nematodes, or other helminths have on cockroaches. Possibly pathogenic microorganisms can be used for biological control of cockroaches; this approach seems to have been little explored.

Associations of colonial cockroaches (e.g., Cryptocercus spp.) may be truly familial or they may merely result from gregariousness. Newly hatched nymphs of species that carry their oÖthecae until the eggs hatch cluster near the mother. This may be a response to the mother as such, a search for shelter beneath the nearest object (thigmotaxis? or negative phototaxis?), or there may well be yet another explanation. Tepper in 1893 stated that the native cockroaches of Australia are almost wholly carnivorous; little supporting evidence for this claim has been brought forward since that time. The apparent supersedure of one species of domiciliary cockroach by another may result from antagonism between different species, or it may result from more rapid breeding and more effective utilization of available food and space; but which? Several species of cockroaches are frequently found associated with certain plants (e.g., bromeliads and bananas); the ecological relations in these associations remain to be determined. Many of the obscure associations between cockroaches and other insects, spiders, birds, and burrowing animals have never been adequately defined. The factors influencing cannibalism have never been thoroughly investigated experimentally. These are only a few ideas for future work that have occurred to us during the preparation of this review. We hope that these suggestions as well as other questions that may occur to readers will stimulate further research in areas where it is obviously needed.

ILLUSTRATIONS

Unless otherwise credited, the illustrations were prepared from photographs taken by the authors. Except where otherwise stated, all photographs were taken of unposed living specimens.