The ecology of extinct cockroaches is necessarily a highly speculative subject. From the coexistence of fossil cockroaches and fossil plants in the same geological stratum, one might conclude that there had been intimate associations between them during prehistoric life. Heer (1864) and Goldenberg (1877) suggested that Carboniferous cockroaches fed on the plants with which they have been found as fossils. Scudder (1879) concurred with this hypothesis. However, Bolton (1911), remarking on the noticeable associations of blattoid wings with vegetable remains, suggested that the cockroaches may have been partly carnivorous, feeding on the snails Spirorbis pusillus, which were attached to the leaves of Cordaites. Yet the proximity of fossil insects and plants in the same geological formation is hardly proof of a similar association during life. In fact, Sellards (1903), Bolton (1921), and Laurentiaux (1951) have all pointed out that the cockroach remains, particularly the more resistant wings, may have been washed into streams by heavy rains and transported with drifting plant material to places where permanent deposits were accumulating.

Some species of fossil cockroaches have long, well-developed ovipositors, very unlike present-day cockroaches whose ovipositors are small and nonprotruding. Brongniart (1889) and Zalesskii (1939, 1953) have suggested that certain Permian and Carboniferous cockroaches with long ovipositors may have inserted their eggs singly into trees and other plants, rather than protecting the eggs with an oÖtheca. However, Laurentiaux (1951), although conceding the possibility of egg laying in vegetable material, suggested that oviposition into the earth is more probable because of the unbending nature of the ovipositor.

Although the ecological associations of modern cockroaches should be well known from direct observation, actually most species are still little more than names on museum specimens, and our knowledge of them is fragmentary. All too frequently ecological observations have been only incidental to taxonomic or faunistic studies; yet the biological information that is contained in such papers is all that we know of many species. For this reason we have cited these observations in some detail, especially when they were brief; longer accounts of cockroach bionomics, of necessity, have been abstracted.

Very few exclusively ecological studies of insects have included cockroaches. The native woodroaches (Parcoblatta pensylvanica, P. uhleriana, and P. virginica) of the northern United States were included in ecological studies of the Orthoptera by Hubbell (1922), Strohecker (1937), and Cantrall (1943). Fifteen species of cockroaches were included in an ecological study of the Orthoptera of northern Florida by Friauf (1953). The original papers should be consulted for detailed descriptions of the habitats and accounts of the associated plants and other Orthoptera.

In this chapter the cockroaches are grouped into those that have been found in man-made structures and those that occur in other habitats. Certain species may appear in several categories because they live both indoors and out. The structural pests are divided into cockroaches that occur in land-based structures, those on ships, and those in aircraft. The nonstructural cockroaches are divided into those that occur in quite specific habitats (caves, water, and deserts) and those that occur generally out of doors. Nests of various arthropods serve as microhabitats of commensal cockroaches; these latter associations are discussed on pages 310-318.

In this chapter our discussion is limited to the physical environment and specific habitats of cockroaches, and only very general references are made to associated organisms. The relationships of cockroaches to the biota are examined in detail in subsequent chapters. To show the full extent of the associations, the associates, from bacteroids to vertebrates, are arranged phyletically. These associate-centered classifications serve admirably to relate various species of cockroaches within common bounds, but fail to give an integrated account of the total biotic relationships in the ecology of each species. Although physically separated in this monograph, the many associates of each species of cockroach should all be considered in appraising the ecology of that species. To assist the reader to achieve this end, we have included a checklist (p. 290) which serves as a convenient index to certain organisms associated with particular species of cockroaches.

CAVE HABITATS

Caves, mines, and animal burrows are somewhat similar habitats that provide many species of cockroaches with shelter and frequently with food. The microclimates of these cockroach habitats have not been described in detail in the papers cited, but it seems rather obvious that natural caves, man-made caves (mines), and burrows offer relatively stable temperatures and humidities and protection from adverse climatic conditions. Although such cavernicolous animals as birds and bats periodically leave caves to search for food, cockroaches find the accumulated guano and animal and plant detritus an entirely adequate dietary (Chopard, 1938). Cockroaches in mines presumably subsist on the food and feces dropped by man and mine animals (e.g., pit ponies). Food stored in their nests by burrowing animals is undoubtedly utilized by the associated cockroaches.

Cavernicolous cockroaches show varying degrees of dependence on and adaptation to these specialized habitats. Some of the common domiciliary species (Blatta orientalis, Blattella germanica, and Periplaneta americana) may have accompanied man into caves and remained there after he left (Chopard, 1929a, 1936, 1938). Other species, from the paucity of records noting their occurrence in caves, are undoubtedly accidental inhabitants that may never become established. Besides these, however, many other species of cockroaches have established large breeding colonies in caves. Although some of the latter species show very pronounced morphological adaptations to a cave life, many others resemble their noncavernicolous relatives. The possible origin of cavernicolous Orthoptera has been discussed by Chopard (1938).

Cavernicolous cockroaches have been segregated into four groups according to their ability to adapt to their environment and the degree of their specialized evolution (Chopard, 1936, 1938): (1) Trogloxenes: Cockroaches that occur in caves in a sporadic fashion (the domiciliary cockroaches and accidentals such as Ectobius and Heterogamodes). (2) Troglophiles: Cockroaches found habitually in caves (Symploce, Periplaneta cavernicola). (3) Guanobies: Cockroaches that live in the guano of cavernicolous vertebrates (Gyna, Acanthogyna, Dyscologamia, Pycnoscelus). (4) Troglobies: Cockroaches that apparently cannot live outside of caves and which show very marked adaptive characters (Alluaudellina, Nocticola, Spelaeoblatta, Typhloblatta). For complete discussions of these groups including descriptions of the adaptive characters shown by certain genera, the original sources should be consulted.

Although we know very little of the ethology of most of the cavernicolous cockroaches, it is intriguing that three of the six known species of Nocticola are cave dwellers, two are inhabitants of termite nests (p. 315), and one (N. bolivari) was found under stones and cement blocks (Chopard, 1950b). In the rather extensive list of cavernicolous cockroaches only two (Arenivaga grata and Parcoblatta sp.) were taken from caves in North America north of Yucatan. All other records are from Africa, Asia, Central America, Europe, West Indies, East Indies, and the Philippine Islands. This we find puzzling. Packard (1888) in his extensive study of the cave fauna of North America listed no cockroaches. Dearolf (1941) found only the above-mentioned Parcoblatta in one of 37 caves in Pennsylvania. Kohls and Jellison (1948) listed no cockroaches among the arthropods from six bat caves in Texas. We would expect Periplaneta americana to inhabit mines in North America, but we have found no such records. Have cockroaches been ignored in fauna collections from North American caves, or has our cave fauna been less extensively studied than that of other parts of the world?

The two species of cockroaches found in mines (Blattella germanica and Periplaneta americana) are also found in caves. For this reason we have included them in the list headed Cavernicolous Cockroaches. On the other hand, the cockroaches found in animal burrows are generally different species from those found in caves, so we have grouped these together in a second list.

CAVERNICOLOUS COCKROACHES

Alluaudellina cavernicola

Tanganyika.—From Kulumusi caves, near Tanga. The eyes of this cockroach are reduced to a pair of slender streaks (Shelford, 1910a; Chopard, 1932a).

East Africa.—Chopard (1936).

Apotrogia angolensis

Belgian Congo.—A troglophile without well-marked adaptive characters. Collected in moist sand on floor of a sandstone grotto inhabited by bats (Chopard, 1927, 1950a). Taken in many caves in Bas Congo (Leleup, 1956).

Apteroblatta perplexa

East Africa.—Accidental inhabitant of cave (Chopard, 1936).

Arenivaga grata

Arizona.—"A female and many nymphs were taken by Flock in the guano in a bat cave in the Tucson Mountains" (Ball et al., 1942).

Aspiduchus borinquen

Puerto Rico.—In limestone cavern by thousands in grass and on walls (Rehn and Hebard, 1927; Rehn, J. W. H., 1951a).

Aspiduchus cavernicola

Puerto Rico.—In limestone cave, in caves inhabited by bats, and apparently seen in other caves well removed from entrance. "In this latter situation great numbers were seen on the side walls and roof" (Rehn, J. W. H., 1951a).

Blaberus atropos

Yucatan.—Found once, in Xmahit cave (Pearse, 1938).

Blaberus craniifer

Yucatan.—Collected within three caves, near the entrances (Pearse, 1938).

Blaberus giganteus

Panama.—Two males and several nymphs were taken under rocks in the second chamber of the Chilibrillo cave; some also were on the walls (Caudell, 1924).

Blatta lateralis

Turkmen S.S.R.—All stages, but more often females and nymphs, were found in the middle and back part of Bakharden cavern, which was inhabited by tens of thousands of bats (Vlasov, 1929).

Blatta orientalis

Turkmen S.S.R.—All stages found in front part of Bakharden bat cave. This cave was uninhabited by man but supported a variety of other animals (Vlasov, 1929).

Blattella germanica

South Africa.—Numerous in a gold mine on the Witwatersrand (Porter, 1930).

Tonkin.—Chopard (1929a); Colani (1952).

Byrsotria fumigata

Cuba.—Cueva de las Cucarachas, La Pantana, Baracoa, Oriente Province: 21 specimens, "It is evident ... that the species is also a cave inhabitant" (Rehn and Hebard, 1927).

Deropeltis erythropeza

East Africa.—Found at entrance of cave; not a strictly cavernicolous form according to Chopard (1936).

Ectobius pallidus

France.—Nymph in cave in Basses-PyrÉnÉes, accidental inhabitant (Chopard, 1936).

Ectobius vittiventer

Italy.—In detritus at base of entrance shaft of Acquaviva cave in the Venezia Tridentina (Conci, 1951).

Ectobiussp.

Italy.—Found in the heap of saprophytic detritus at the base of the entrance shaft in the Acquaviva cave (Conci, 1951).

Ergaula scarabaeoides

Sumatra.—West coast (Hebard, 1929).

Malaya.—Found burrowing in bat guano among stones at entrance to caves in Selangor (Chopard, 1919, 1929).

Euthyrrhapha nigra

Madagascar.—Three males and six females in guano in Antsinomy grotto (Chopard, 1949a).

Gyna kazungulana

East Africa.—This species is especially found in caves although it shows no special adaptive characters. It is a typical guanobe (Chopard, 1936).

Gyna maculipennis

Belgian Congo.—Troglophile, guanophile. Found in two caves in Lualaba (Leleup, 1956).

Gyna tristis

Belgian Congo.—In three caves in Uele (Leleup, 1956).

Heterogamodes krÜgeri

North Africa.—An accidental inhabitant of caves (Chopard, 1938).

Holocompsa zapoteca

Yucatan.—Common throughout rather dry, dusty caves in southern Yucatan (Pearse, 1938).

Hoplosphoropyga babaulti

Stated to be a troglophile by Chopard (1938).

Nocticola caeca

Philippine Islands.—BolÍvar (1892).

Nocticola decaryi

Madagascar.—A true troglobite according to Chopard (1945).

Nocticola simoni

Philippine Islands.—BolÍvar (1892).

Parcoblattasp.

Pennsylvania.—Found in Merkle cave, Berks County (Dearolf, 1941).

Periplaneta americana

IN CAVES

East Africa.—Its presence in the cave at Shimoni was thought to indicate that man had sought refuge there and brought the cockroaches in with baggage or provisions (Chopard, 1936).

India.—Many present in cave at Vengurla, the floor of which was covered with bird guano (Abdulali, 1942).

Madagascar.—Thought to have been introduced into the cave entrance by man (Chopard, 1945, 1949a).

IN MINES

Great Britain.—In a coal mine at Pontewydd where they had been established for some years (Lucas, 1916). In the Pentre Pit mine where they were abundant (Lucas, 1918). Abundant in a Welch mine 2,166 feet below the surface (Lucas, 1925). This species was found quite commonly in a number of South Wales coal mines; in one deep mine a white-eyed mutant form comprised about 5 percent of the cockroach population for the preceding 11 years (Jefferson, 1958).

India, western Bengal.—Very numerous in coal mines where the sole food apparently was human faeces (Chandler, 1926).

South Africa.—Numerous in four deep-level gold mines on the Witwatersrand.

Sumatra.—Numerous males and females from Sawah Lunto "'from a coal mine where they lived in great numbers on the faeces of miners'" (Hanitsch, 1929).

Periplaneta australasiae

Sarawak.—Found swarming on walls of caves and in soft bird guano in company with Symploce cavernicola (Moulton, 1912).

Tonkin.—Chopard (1929a); Colani (1952).

Periplaneta cavernicola

Malaya.—Taken on walls of inner caverns, where they were particularly abundant (Chopard, 1919).

Periplaneta lata

Tonkin.—Chopard suggested that its presence in caves is probably linked with man (Chopard, 1929a; Colani, 1952).

Periplanetasp.

Malaya.—From a cave in Jalor (Annandale et al., 1913).

Perisphaerussp.

Malaya.—The wingless females and nymphs mined in bats' guano in a cavern of the Jalor caves (Annandale, 1900).

Polyphaga aegyptiaca

Turkmen S.S.R.—Females found in front part of Bakharden bat cave on several occasions (Vlasov, 1929).

Turkey.—At Magharadjik and Arab Dede, found in caves with various other animals (Lindberg, 1954).

Polyphagasp.

Burma.—Hsin Dawng Cave, S. Shan States, 1 immature male under stone in complete darkness (Chopard, 1924b).

Pycnoscelus niger

Tonkin.—Apparently not an accidental inhabitant as nymphs were present (Chopard, 1929a; Colani, 1952).

Pycnoscelus striatus

Malaya.—Found burrowing in bats' guano at entrance to caves in Selangor, where it was very abundant 50 to 600 feet from entrance; also on walls of inner cavern (Chopard, 1919, 1929). In the absence of other evidence, the presence of P. striatus in a cave indicates that bats also inhabit the cave (Chopard, 1929a).

Pycnoscelus surinamensis

Assam.—Found 300 to 400 feet from entrance of Siju cave in the Garo Hills (Chopard, 1924b).

South Celebes.—Hanitsch (1932).

Spelaeoblatta gestroi

Burma.—Chopard stated that this species shows marked characteristics of adaptation to a life in darkness (BolÍvar, 1897; Annandale, 1913; Chopard, 1919).

Symploce breviramis

South Celebes.—Hanitsch (1932).

Symploce cavernicola

Sarawak, Borneo.—Swarming on walls of caves and in soft bird guano on the cave floor (Moulton, 1912). Hanitsch (1931) noted that this species was first recorded by Shelford from a cave in Sarawak and that there is a series from a cave in the Oxford University Museum, taken by Banks in 1928.

Malaya.—On the walls of the inner cavern of a cave at Biserat; the insects covered the walls in places (Chopard, 1919).

Sumatra.—From Baso cavern, on the west coast (Hebard, 1929).

Symploce curta

South Celebes.—Hanitsch (1932).

Symploce remyi

Tonkin.—This seems to be a true cavernicolous species (Chopard, 1929a; Colani, 1952).

Tivia macracantha

Belgian Congo.—A troglophile without well-marked adaptive characters (Chopard, 1950a). At Haut-Katanga, troglophile and guanophile (Leleup, 1956).

Tivia sp.

Madagascar.—Last-stage nymphs captured in guano in Antsinomy grotto (Chopard, 1949a).

Typhloblatta caeca

India, Assam.—An eyeless species with noticeably elongated appendages (Chopard, 1945).

Typhloblattodes madecassus

Madagascar.—Unpigmented integument and reduced eyes (Chopard, 1945).

Xestoblatta immaculata

Panama.—Found under rocks on guano-covered floor of the Chilibrillo bat caves (Caudell, 1924).

Unidentified cockroaches

Malaya.—The walls of a cave were covered by dense groups of a species of "Blatta" (Annandale, 1900).

England.—"The chief insect pests of the mines are cockroaches, which often swarm in hot mines and those with pit pony stables...." (Hardy, 1941).

COCKROACHES FROM THE BURROWS OF VERTEBRATES

Arenivaga apacha

Arizona.—In the nests of wood rats, Neotoma sp. (Hebard, 1917).

Arenivaga bolliana

Texas.—In the nests of wood rats, Neotoma sp. (Hebard, 1917; 1943a).

Arenivaga erratica

Arizona.—The wingless females were commonly found in burrows of Dipodomys spectabilis spectabilis Merriam, the kangaroo rat. The winged males were never found in the burrows (Vorhies and Taylor, 1922). Found most commonly in wood-rat and ground-squirrel dens in the desert regions (Ball et al., 1942).

Arenivaga floridensis

Florida.—Found in a burrow of Peromyscus polionotus rhoadsi (Bangs), the white-footed mouse (Young, 1949).

Arenivaga roseni

Turkmen S.S.R.—Occasionally found in burrows of Rhombomys opimus Lichtenstein; in the burrows of the desert turtle, Testudo horsfieldi Gray; and frequently in burrows of the ground squirrel, Spermophilopsis leptodactylus Lichtenstein (Vlasov, 1933; Vlasov and Miram, 1937).

Arenivaga tonkawa

Texas.—An immature specimen was found in a prairie-dog hole (Hebard, 1943a).

Cariblatta lutea

Florida.—It has been taken in burrows of the pocket gopher, Geomys sp. (Hubbell and Goff, 1940).

Euthlastoblatta abortiva

Texas.—In the nests of wood rats, Neotoma sp. (Hebard, 1917).

Parcoblatta fulvescens

Texas.—In the nests of wood rats, Neotoma sp. (Hebard, 1917).

Polyphaga aegyptiaca

Turkmen S.S.R.—Nymphs and adult females were often found in burrows of the sand mouse, Rhombomys opimus (Vlasov, 1933).

Polyphaga indica

Turkmen S.S.R.—This species prefers sandy soils where it can be found in burrows of Spermophilopsis leptodactylus and Pallasiomys meridionalis pennicilliger Heptner (Vlasov and Miram, 1937).

Polyphaga saussurei

Tadzhikistan.—Found in burrows of turtles and rodents (Zmeev, 1936).

Turkmen S.S.R.—Nymphs and adult females are common in burrows of Rhombomys opimus and in burrows of Testudo horsfieldi. Its principal habitat is rodent burrows in loess dust, where it is not infrequently found in the food stores of the host (Vlasov and Miram, 1937).

Pycnoscelus surinamensis

Texas.—In the nests of wood rats, Neotoma sp. (Hebard, 1917).

DESERT HABITATS

There is relatively little ecological information about cockroaches that live in deserts, even though certain species, notably Polyphaga aegyptiaca, have long been known to inhabit arid zones. In fact, so little is known about the ecology of arid-zone insects in general that it is more a subject for research than for review (Pradhan, 1957). In their account of the cockroaches of Northern Kenya and Jubaland, Kevan and Chopard (1954) describe in some detail the vegetational areas of this arid desert or semidesert country, which averages only about 10 inches of rain per year. The other sources that are cited below contain very little more biological information than the abstracted material that is given under each species.

Nearly all the Polyphaginae are said to be marked xerophiles whose distribution coincides with that of the deserts (Bei-Bienko, 1950). With the exception of Arenivaga floridana, the species of Polyphaginae in the United States all occur in the Southwest, where they are (with a few exceptions) the only cockroaches that inhabit the desert regions proper (Hebard, 1917). The Polyphaginae reach their greatest diversity in the deserts of Northern Africa and Anterior and South-Central Asia (Bei-Bienko, 1950). Some of the desert-inhabiting species have also been found under nondesert conditions. This only further exemplifies the plasticity of cockroaches in adapting to different environments.

The ability of desert insects to live under what appear to be extremely unfavorable conditions has been abundantly illustrated by Pradhan (1957). Uvarov (1954) has pointed out that a desert "covers a great variety of landscapes, which provide desert animals with a wide range of habitats, some of them offering very favorable conditions for life." Pradhan (1957) stated that many desert animals avoid the extremes of desert climates by choosing suitable microclimates for diurnal resting places, that a permanent or temporary underground existence is very common among insects in arid zones, and that many nocturnal Orthoptera burrow into the soil or hide under stones where temperatures are lower. For example, the type of Parcoblatta desertae was found under a boulder on the bare desert (Rehn and Hebard, 1909).

Symbiosis with burrowing animals is another solution to the problem of existence in the desert; in fact, symbiosis is a mode of life adopted by nearly half of the desert cockroaches about which we have any information. Vlasov and Miram (1937) found Polyphaga indica, Polyphaga saussurei, and Arenivaga roseni in the burrows of rodents and desert turtles. In the desert regions of Arizona, females of Arenivaga erratica were found commonly in burrows of the kangaroo rat (Vorhies and Taylor, 1922) and in dens of wood rats and ground squirrels (Ball et al., 1942). Arenivaga apacha and Arenivaga bolliana have also been found inhabiting the nests of wood rats (Hebard, 1917; 1943a). Bei-Bienko (1950) has suggested that the adaptation of desert-inhabiting cockroaches to rodent burrows might enable these insects to survive in the severe climatic conditions of deserts in summer.

Under desert conditions in southern Arizona, the relative humidity outside of the burrows of the kangaroo rat is 1 to 15 percent during the day and 15 to 40 percent at night; but inside the burrows the relative humidity is 30 to 50 percent, and the temperature, even during the day, is below 30° C. (Schmidt-Nielsen, 1949). Thus by living in rodent burrows during the day and going outside at night, the desert cockroaches could avail themselves of the most favorable microclimates obtainable. Presumably whatever food these insects eat provides them with sufficient water to enable them to survive under desert conditions. Bodenheimer (1953) has suggested that the extent of utilization of dew, which is sometimes heavy in the desert, should be investigated; he stated that tenebrionid beetles have been seen in the early morning eating dry [dead?] herbs that were still wet with dew. It is obvious that there is a need for additional detailed information without which we can only guess about the ecology of desert cockroaches.

In the following list we have cited only those species that were stated to have been found under desert conditions. Undoubtedly, related species that have been taken in similar localities are also desert-inhabiting forms, as, for example, other species of Arenivaga that were collected in Texas by Hebard (1943a). In the absence of specific information linking such other species with deserts, we have arbitrarily relegated those forms to the section on outdoor habitats. In addition to the species listed below, desert cockroaches are said to be found in the following genera: Anisogamia, Mononychoblatta, and Nymphytria (Chopard, 1938).

DESERT COCKROACHES

Agis orientalis

Northern Kenya.—In desert-grass and thorn-bush country; scattered, dry tufts of grasses interspersed among acacia bush and scattered trees (Kevan and Chopard, 1954).

Arenivaga apacha

U.S.A.—Inhabits desert regions of the Southwest, has been found in nests of wood rats (Hebard, 1917).

Arenivaga bolliana

U.S.A.—On gravelly hillocks, in scattered scrub, and in the nests of wood rats in Texas. It is a desert inhabitant in the Southwest (Hebard, 1917; 1943a).

Arenivaga erratica

U.S.A.—Inhabits desert regions of the Southwest (Hebard, 1917). In Arizona it has been found in rodent burrows in the desert (Vorhies and Taylor, 1922; Ball et al., 1942).

Arenivaga roseni

Turkmen S.S.R.—Predominantly found in burrows in sand; all stages "swim" in sand and loess dust (Vlasov and Miram, 1937).

Blattella vaga

Arizona.—Found in small numbers on the dry desert (Flock, 1941a).

Compsodes schwarzi

U.S.A.—Occurs in the Southwest where it is confined to the desert and semidesert mountainous areas, rarely being found on the desert floor (Hebard, 1917). Taken in an ant nest in mountains of Arizona (Ball et al., 1942).

Cyrtotria capucina

Eastern Africa.—"Commonly met with under dÉbris, the apterous females being most frequent." Thorn-bush country (Kevan and Chopard, 1954).

Derocalymma lampyrina

Northern Kenya.—Very abundant; both sexes under dÉbris in desert-grass and thorn-bush country (Kevan and Chopard, 1954).

Derocalymma porcellio

Northern Kenya.—Taken in upland grassland and bush (Kevan and Chopard, 1954).

Deropeltis autraniana

Northern Kenya.—In thorn-bush country (Kevan and Chopard, 1954).

Deropeltis melanophila

Northern Kenya.—"Very commonly found at the base of tufts of grass and other dÉbris, the apterous female particularly in the latter situation"; in upland grassland near forest; in thorn-bush country (Kevan and Chopard, 1954).

Deropeltis nigrita

Northern Kenya.—Taken in upland grassland and bush (Kevan and Chopard, 1954).

Eremoblatta subdiaphana

U.S.A.—Apparently found in greatest abundance in the extreme desert conditions of the southwestern United States (Hebard, 1917). Two small groups of males were observed in the midst of the sandy desert north of Yuma, Ariz.; these insects alternately flew and ran over the sand in the hot sun while headed in a southwesterly direction (Wheeler, 1911).

Euandroblatta palpalis

Northern Kenya.—In desert-grass and thorn-bush country (Kevan and Chopard, 1954).

Heterogamodes rugosa

Northern Kenya.—"All from desert grass and thorn bush (on sand)." It was stated (under discussion of Tivia fulva) that Heterogamodes females live more or less buried in the sand (Kevan and Chopard, 1954).

Namablatta bitaeniata

Southwestern Africa.—Limited in distribution to the more arid portions, being peculiar to extreme desert conditions (Rehn, 1937).

Nauphoeta punctipennis

Northern Kenya.—In desert grass and thorn bush; "probably the commonest of all the medium-sized cockroaches occurring in the area under discussion, coming very freely to light" (Kevan and Chopard, 1954).

Parcoblatta desertae

U.S.A.—In the desert and semidesert mountainous areas of the Southwest; it is rarely found on the desert floor (Hebard, 1917). Found under boulder on bare desert (Rehn and Hebard, 1909).

Polyphaga aegyptiaca

Caucasus.—The wingless female was found buried in sand and dust (Burr, 1913).

Turkmen S.S.R.—Although this species is secondarily encountered in dwellings and courtyards, it is a very characteristic insect of the Trans-Caspian deserts; the females are encountered fairly frequently as inhabitants of sand, where they run slowly over the surface, or dig themselves into the sand to continue their forward motion not far below the surface (Fausek, 1906). Uvarov (in Chopard, 1929b) indicated that females of this genus are found in various desert localities, particularly where vegetative debris occurs, but they are not strictly attached to sandy terrain.

Polyphaga indica

Turkmen S.S.R.—This species prefers sandy soils where the nymphs, alate males, and wingless females "swim" readily through the sand; they can also be found in the burrows of desert animals (Vlasov and Miram, 1937).

Polyphaga saussurei

Turkmen S.S.R.—Its principal habitats are rodent burrows in loess dust and burrows of the desert turtle (Vlasov and Miram, 1937).

Supella hottentotta

Northern Kenya.—Taken in bushes by dry river bed and in desert-grass and thorn-bush country at several stations (Kevan and Chopard, 1954). " ... taken with light at night running on bark of a large acacia tree" (Rehn, 1947).

Symploce kevani

Northern Kenya.—In desert grass and thorn-bush country (Kevan and Chopard, 1954).

Theganopteryx straminea

Northern Kenya.—Taken at three stations in desert grass and thorn bush (Kevan and Chopard, 1954).

Tivia brunnea

Northern Kenya.—In open sandy, riverine bush (scanty ground cover among acacia trees and doum palms) (Kevan and Chopard, 1954).

Tivia fulva

Northern Kenya.—In desert grass and thorn bush; distributed in semidesert areas south of Sahara; the apterous females probably live buried in sand (Kevan and Chopard, 1954).

Tivia obscura

Northern Kenya.—In desert grass and thorn bush (Kevan and Chopard, 1954).

AQUATIC HABITATS

The so-called aquatic or amphibious cockroaches are all members of the subfamily Epilamprinae (Chopard, 1938). These forms are not nearly as aquatic as water beetles or aquatic Hemiptera, but in their relations to water they behave differently from nonamphibious cockroaches, which tend to avoid water except for drinking. There are apparently no special morphological characteristics that distinguish amphibious cockroaches (Shelford, 1907, 1909a; Chopard, 1938), although Takahashi (1926) listed several characters that he considered made Opisthoplatia maculata adapted for an aquatic life: (1) Back of body easily wetted; (2) long hairs on underside of thorax trap air; (3) terminal abdominal spiracles open into tubes that extend rearward; (4) long hairs on ventral surfaces of cerci "protect" terminal abdominal spiracles. Annandale (1906) also suggested that the position of the posterior abdominal spiracles, at the base of tubes that project rearward from beneath the seventh tergite, are an adaptation to an aquatic life. However, as Shelford (1907) and Chopard (1938) have pointed out, this same feature may be observed in many terrestrial cockroaches. The legs of amphibious cockroaches are similar to those of nonaquatic species and are not modified for swimming (Shelford, 1909a; Takahashi, 1926).

Biological observations have been made on relatively few species, but representatives of at least six genera occur in quasi-aquatic habitats. Strictly speaking, these cockroaches live on land at the edges of streams or pools and spend relatively brief periods in the water. A few species are found in water-filled bromeliads. The behavior of the known amphibious species of cockroaches in relation to their habitats is discussed below.

AMPHIBIOUS COCKROACHES

Audreia bromeliadarum

Panama.—These insects when disturbed would dive into the water that had collected in the base of the bromeliad; they would disappear beneath the surface and remain submerged for some considerable time (Caudell, 1914).

Dryadoblatta scotti

Trinidad.—This species was taken from the leaf bases of Tillandsia sp. at 3,100 feet; water had collected between the leaves and the insect was presumed to be more or less amphibious (Scott, 1912). Subaquatic in the bromeliad Glomeropitcairnia erectiflora: "This large and handsome species [D. scotti] is very common in the larger, water-filled, epiphytic bromeliads of the rain forest. Within these plants it is usually to be found, often in considerable numbers, just above the surface of the water or partly immersed in it. The cockroaches will descend rapidly into the water when alarmed and probably obtain their nourishment from the accumulated organic matter in the water. Floating material is probably taken and it seems less likely that they feed below the surface. They appear to be ovoviviparous." (Princis and Kevan, 1955.)

Epilampra abdomen-nigrum

Puerto Rico.—Abundant in wet "malojillo" meadows. The nymphs swim easily and remain under water for long periods, as do the adults (SeÍn, 1923; Wolcott, 1950).

Panama.—A swimming nymph, captured in a dipper with mosquito larvae in a lagoon of the Rio Chilibre, was kept under observations in an aquarium. If disturbed, the insect dived into the water from floating vegetation and swam rapidly below the surface for a minute or two. Finally becoming quiescent, the cockroach would then cling to submerged roots; twice it remained still for 15 minutes before climbing to the surface, where it remained for five or more minutes before emerging completely (Crowell, 1946).

Epilampra annandalei

Lower Burma.—One male and three nymphs were collected in the Dawna Hills by Annandale who made the following observations: "The wingless specimens were under stones in a jungle stream and behaved just as the one I obtained in Chota, Nagpur, did [Annandale, 1906]. The winged specimen was under a stone at the edge of the stream, but swam readily. It did not seem so much at home in the water, however, and apparently could not, owing to the wings, raise the tip of its abdomen above the surface." (Shelford, 1909a.)

Epilamprasp.

Siamese Malay States.—Wingless females rested on floating logs from which they would dive into the water upon the least disturbance; they remained under water for several minutes, then surfaced beneath the shelter of the log. In the jungle all females were taken either in the water or among matted roots on the sides of the stream. Winged males were seen rising from the surface of the water (Annandale, 1900).

Sarawak.—All specimens were immature; they swam and dived well, but were soon drowned if prevented from rising to the surface to breathe. "When at rest the body of the cockroach is almost entirely submerged, the tip of the abdomen alone projecting above the surface of the water; the abdomen moves gently up and down and every 30-40 seconds a bubble of air issues from the prothoracic spiracle on each side." (Shelford, 1901, 1916.)

India.—A nymphal female, found in a jungle stream at Chota Nagpur, could swim with belly or back upward. When held under water it drowned in a few minutes. The tip of the abdomen was held out of water (Annandale, 1906).

Shelford (1907) has suggested that the immature stages of terrestrial species of Epilampra may well be amphibious. This is an area that could profit by more field observations.

Opisthoplatia maculata

Formosa.—Invariably found under or between rocks near mountain streams. The wingless adult and the nymph have similar habits. Normally the cockroach lives on land, and when it goes into the water it returns to land within a few minutes. This cockroach rarely swims, but when it does, it maintains its body in a horizontal position just below the surface of the water. Ordinarily, it walks on the river bottom or on water-covered rocks. This insect feeds on decayed leaves and, according to Shikano, it will eat human feces. (Takahashi, 1926.)

This species has a large number of long hydrophobic hairs on the ventral sides of the thorax and anterior abdominal segments. When the insect submerges, air is trapped in these hairs. The thoracic and one pair of abdominal spiracles open into the bubble of trapped air. However, the insect apparently does not use this plastron of air to replenish its tracheal air supply, but, like Rhicnoda natatrix (see below), it inspires air while at the surface through its posterior abdominal spiracles and expires air into the bubble under the thorax. While the insect is submerged, the air bubble increases in volume until part of it breaks away and floats to the surface. (Takahashi, 1926.)

Opisthoplatia orientalis

Formosa.—Lives on or in swampy ground (Takahashi, 1924).

Rhicnoda natatrix

Sarawak.—Immature cockroaches were found in sodden leaves at the edge of a pool, where they rested for hours at a time. Generally the fore part of the body was in the water but the tip of the abdomen was always in air. When disturbed the insects dived into the water and hid under sticks and stones on the bottom. Air is inspired through the posterior abdominal spiracles, when they projected above the water surface, and expired through the thoracic spiracles. In experiments in which the insects' abdomens were held immersed in water, with the thorax exposed, the insects died in 6 to 12 hours or less. (Shelford, 1907.)

Stictolampra buqueti concinula

Westsumba.—Found under moist fallen leaves on gravelly shore of Melolo River. The nymphs distinguished themselves through their amphibious mode of life and were often good swimmers (Princis, 1957a).

Unidentified epilamprines

Brazil.—These cockroaches were found under stones at the side of a rocky stream at Ouro Preto. When disturbed they ran down under the surface of the water and hid under stones at the bottom. When thrown on the water surface, they were helpless, and to get beneath the water surface they had to walk down some object. When they had penetrated the surface film they could swim freely. Specimens kept in jars lived several days with only a portion of their abdomens exposed to the air. (Bristowe, 1925.)

OUTDOOR HABITATS

This category is a catchall for all cockroaches that are not limited to the more circumscribed habitats that have been previously considered. Some cockroaches in this section select specific microhabitats (e.g., Cryptocercus spp., which live exclusively in rotten logs; and Neoblattella dryas, N. eurydice, and N. grossbecki in bromeliads). Others are found in a wide variety of habitats (e.g., Ischnoptera deropeltiformis and Parcoblatta spp.). But some species are so little known that their actual habitats are barely suggested in the collection data.

Williams (1941) made an ecological study of the floor fauna of the Panama rain forest. He found Orthoptera (nearly all were unidentified nymphal cockroaches) in the litter of dead leaves, twigs, and other plant products in over 90 percent of the quadrats he examined. These insects represented about 0.25 percent of the total animal population.

Delamare Deboutteville (1948) made a quantitative study of the animal population in suspended soil that had accumulated between the roots of forest epiphytes of the lower Ivory Coast. He analyzed 2 dm.³ samples of soil from an epiphyte located 45 meters above ground on a main branch of Parinarium, with these results: Horizon A.—Superficial zone of large rootlets, 6 cm. deep: 2 cockroaches, 4 arachnids, and 4 beetles. Horizon B.—Zone of fine rootlets, 6 cm. deep: 6 cockroaches and numerous other arthropods. Horizon C.—Humid zone, 8 cm. deep: 7 cockroaches and numerous other arthropods. Plants, such as Palissota, were also living in this very original biotype.

The species of cockroaches listed below have been found in the following kinds of outdoor microhabitats: In jungle, forest, and woodlands they have been found in rotten wood; under bark of living, dead, and fallen trees; in decay cavities in trees; burrowing in living bark; on foliage of trees, shrubs, bushes, and low herbage; on vines and in bromeliads and epiphytic ferns; under signs on trees and stumps; in piles of logs and firewood; under dead leaves and debris; in and under decaying fruit on the ground. Cockroaches have been found between the leaves and under leaf sheaths of sugarcane, corn, and other grasses; under dry fibers and fronds of coconut trees; in hollow stems and bases of tree-fern fronds; under bracts of banana blossoms and in bunches of bananas (p. 146). Cockroaches also inhabit abandoned cocoons and larval tents, wasp nests, ant nests, termite nests, bird nests, rat nests, and burrows of other rodents (pp. 23-25, 310-319). Cockroaches have been found in rock crevices and under rocks; under boards and other objects on ground; under seaweed, drift, and other debris on beaches; burrowing in soil and under clods of earth; in marshes and swamps; in dumps and rubbish heaps.

The above list does not exhaust the available outdoor microhabitats that cockroaches find suitable for their continued existence, but it is fairly representative. Although we have no measurements to substantiate this conclusion, we suggest that the microhabitats cited above have a more constant temperature and a relatively higher humidity than is provided by the surrounding macrohabitats. We would expect insects such as cockroaches, whose water balance is dependent on a continuous supply of fluid water or moist food, to seek moist environments or to avoid situations in which their transpiration might increase. Deviations, presumably brief, from this expected behavior must occur to account for the cockroaches that are found under relatively unfavorable environmental conditions. Despite the apparent preference for cryptic habitats, some cockroaches are found in hot sunlight (Ellipsidion spp.; Tepper, 1893); Rehn (1945) has stated that many kinds are diurnal rather than nocturnal. Movement of cockroaches between habitats may be assumed to occur; but movement from an unfavorable environment to a more favorable one, following a shift in water balance, has not been observed in nature; however, laboratory experiments suggest that the mechanism for mediating such behavior is present in some species of cockroaches (Gunn and Cosway, 1938; Roth and Willis, 1952a). Obviously, additional research is needed on the bionomics of all species. Further conclusions based on current limited knowledge can only be speculative and possibly misleading.

COCKROACHES FROM OUTDOOR HABITATS

(Except Amphibious, Desert, and Cavernicolous Forms)

Aglaopteryx absimilis

Puerto Rico.—Living in rotten, wooden fence; living between leaves of Samanea saman and in abandoned cocoons of Megalopyge krugii on bucare trees (Wolcott, 1950).

Leeward Islands.—On coconut tree (Princis and Kevan, 1955).

Aglaopteryx facies

Puerto Rico.—As diaphana, in dead branch 10 feet above the ground on Mona Island (Hebard, 1917). In trunks of trees under bark and very often in abandoned cocoons of the "plumilla" (SeÍn, 1923). On rotten, wooden fence; in empty cocoons of Megalopyge krugii on trunks of bucare trees, Erythrina glauca; on trunk of Inga laurina; in larval tents of Tetralopha scabridella on Inga vera (Wolcott, 1936). In large numbers in nests of the gray kingbird (Wolcott, 1950).

Aglaopteryx diaphana

West Indies.—In Cuba, under corky bark of large tree in open; Jamaica, under loose bark of shade trees and in bracts of banana blossoms; in bromeliads and hollow bases of dead tree-fern fronds (Rehn and Hebard, 1927).

Aglaopteryx gemma

Florida.—On Long Key, under coquina boulder in heavy scrub; under loose, dry fibers near head of standing coconut palm (Rehn and Hebard, 1912). Climbing on roots of red mangrove, Rhizophora mangle, in swamp; under loose bark on trunk of Exothea paniculata in dense jungle; under limestone boulder in keys scrub; under signs on oaks, sweet gum, and longleaf pines in southeastern and southern States (Hebard, 1917). Infrequent in the shrub growth of the Sandhills habitat (Friauf, 1953).

Texas.—In undergrowth of pine forest; under sign on oak near river; in Tillandsia sp. (Hebard, 1917). Usually in hiding places on trees; only once found under a stone on ground (Hebard, 1943a).

Allacta similis

Hawaii.—Common in hollow stems and under bark (Swezey and Williams, 1932).

Amazonina emarginata

Trinidad.—On low herbage, on hibiscus at night, and in banana bunch (Princis and Kevan, 1955).

Anaplecta asema

Panama.—Under dead leaves in jungle (Hebard, 1920).

Anaplecta decipiens

Costa Rica.—In decayed leaves (Rehn, 1906).

Anaplecta fallax

Costa Rica.—Under stones on borders of Surubres River (Rehn, 1906).

Anaplecta hemiscotia

Panama.—Under rubbish at edge of jungle and in overgrowth of heavy vines on low bushes (Hebard, 1920).

Anaplecta lateralis

Panama.—Under drift on edge of coral-sand beach (Hebard, 1920).

Arenivaga bolliana

Texas.—In dense jungle brush of the river plain; on gravelly hillocks in scattered scrub; under debris and leaf mold under mesquite trees; in rat's nests, Neotoma sp. (Hebard, 1917). In dry earth under bush; inhabits litter on ground and nests of rats (Hebard, 1943a).

Arenivaga floridensis

Florida.—Male on ground under leaves of cabbage palmetto (Blatchley, 1920). Females in sand under boards and debris along lake shore (Friauf in Cantrall, 1941). Infrequent on bare soil and ground under vegetation in the longleaf-pine flatwoods habitat (Friauf, 1953). In rodent burrow (Young, 1949).

Arenivaga grata

Texas.—Under stones in upper canyon; under rocks in pine-oak forest; from oak-manzanita forest along dry stream bed (Hebard, 1943a).

Aristiger histrio

Malaya.—Lives freely on bushes and flowers of Passiflora sp. (Karny, 1924).

Aspiduchus boriquen

Puerto Rico.—"Apparently the species [as deplanatus] is locally numerous in suitable locations, such as caves, rock crevices and the shelter of large stones." (Rehn and Hebard, 1927).

Audreia bromeliadarum

Panama.—Perfectly at home in bromeliads (see p. 31) (Caudell, 1914).

Audreia jamaicana

Jamaica.—In bromeliads; under dead wood in dense forest (Rehn and Hebard, 1927).

Balta godeffroyi

Australia.—Under bark (Hebard, 1943).

Balta quadricaudata

Australia.—From sugarcane (Hebard, 1943).

Balta scripta

Australia, Queensland.—On leaves, grass, and sugarcane (Hebard, 1943).

Balta torresiana

Australia.—From leaves, under bark, from sugarcane (Hebard, 1943).

Balta verticalis

Australia.—In leaves, from tree, from sugarcane (Hebard, 1943).

Blaberus atropus

Trinidad.—Female in rotting log (Princis and Kevan, 1955).

Blaberus discoidalis

Jamaica.—Under dead coconut petioles in open spot. Gundlach found it under stones in a field in Cuba (Rehn and Hebard, 1927).

Blaberus giganteus

Trinidad.—Nymph in rotten palm tree (Princis and Kevan, 1955).

Blaberusspp.

Venezuela.—Only taken in the forests of the Orinoco near the trunks of rotten trees at night (Doumerc in Blanchard, 1837).

Panama.—Among dead leaves and debris on floor of rain forest (E. C. Williams, Jr., 1941).

Blatta lateralis

U.S.S.R.—Found among rocks at 2,000 or more meters elevation. It is found in cultivated areas as well as in mountainous landscapes and in semideserts (Bei-Bienko, 1950).

Blatta orientalis

Great Britain.—One female nymph under bark of tree 10 feet above the ground (Burr, 1900). Swarming within a rubbish heap in February (Lucas, 1912). In refuse tip under old sacks and sheets of linoleum (Hallett in Lucas, 1922). Male under bark of oak far from houses (Donisthorpe, 1918). One adult female and nymph in prone dead elm 50 yards from house (Burr, 1937). An immature male at the roots of Ballota nigra (Buck in Gardner, 1954). Four additional records of this species outdoors away from houses (Lucas, 1920).

Southern Crimea.—Under stones, dead leaves, and detritus in small copses of Quercus pubescens, Carpinus orientalis, Cornus mas, Paliurus aculeatus, and Dictamnus fraxinella; 19 specimens, apparently breeding outdoors (Adelung, 1907).

North-central U.S.—Observations since 1950 indicate a marked increase in frequency and duration of infestations outdoors; observed in bare soil, vegetation, debris, alongside foundations in sodded areas, along sidewalks, and at edge of parking areas throughout the year; in some urban residential areas, the yards of whole blocks of houses were "alive" with this species on warm summer nights; in winter they have been found under stones, leaf debris, and soil near structures (Shuyler, 1956).

Blattella germanica

Algeria.—Under moist leaves in woods (Lucas, 1849).

California.—Under rubbish and on date palms (Herms, 1926).

Connecticut.—In city dump under loose material, very numerous (Walden, 1922). Additional infestations of dumps by this species have been reported in New York (Felt, 1926, 1928) and New Jersey (Hansens, 1949, 1950)

England.—Swarming within a rubbish heap in February (Lucas, 1912).

Formosa.—Lives among fallen leaves on the ground (Takahashi, 1924).

North-central U.S.—Reported living outdoors near buildings and in soil under basementless buildings from early summer to late fall (Shuyler, 1956).

Blattella humbertiana

India.—Common among decaying vegetation and on trees (Chopard and Chatterjee, 1937).

Formosa.—Normally found in sugarcane fields, pineapple fields, and grasslands where it feeds on decayed leaves and other decayed vegetable matter and dead insects. It lies concealed among and under fallen leaves and clods of earth on or close to ground and never on the upper parts of plants, except pineapple where it is found among the leaves (Takahashi, 1940).

Blattella vaga

Arizona.—Typically an inhabitant of irrigated fields and yards, it is found in fewer numbers on the dry desert. It is found under stones, plant debris, and clumps of earth; found in greatest numbers around decaying dates on ground (Flock, 1941a).

Texas.—Beneath duff under athel trees; rather abundant in clumps of Rhodes grass (Riherd, 1953).

Byrsotria cabrerae

Cuba.—In sea-coast woods: "The species [this and Byrsotria fumigata] are ground-dwelling, hiding under stones and other shelter" (Rehn and Hebard, 1927).

Byrsotria fumigata

Cuba.—Ground dwelling, hiding under stones, etc.; also a cave inhabitant (Rehn and Hebard, 1927).

Cahita borero

Brazil, Matto Grosso.—Beaten from tree foliage in dry scrub, from tree foliage at edge of dry riverine tangle, and from undergrowth in a dry forest area (Rehn, 1937a).

Cahita nahua

Honduras.—All beaten from foliage along roads or in thickets, during rainy season (Rehn, 1937a).

Cariblatta antiguensis

Virgin Islands, St. Croix.—Common under heaps of rubbish (Beatty, 1944).

Trinidad.—On herbage below bananas; all stages on Hibiscus at night; in grass at dusk; on low herbage under old coconut (Princis and Kevan, 1955).

Cariblatta cuprea

Jamaica.—In leaves on leaf mold in hillside forest (Hebard, 1916a).

Cariblatta delicatula

West Indies.—In debris in short grass in open, Cuba. Under dead petioles of coconut palms, San Domingo. In leaves on leaf mold in hillside forest, Jamaica (Hebard, 1916a).

Cariblatta hylaea

Honduras.—Found at foot and on lower slopes of first ridges of the Sierra Pija, from 75 to at least 800 feet above sea level, where vegetation ranged from abandoned banana patches overgrown with Heliconia and Cecropia and interspersed with forest trees, at the foot of the hills, to primeval lowland forest (ceibas, figs, palms, etc.) on the slopes. In the banana patches C. hylaea was found on hanging dead banana and Cecropia leaves; on the slopes it was found on undergrowth foliage, hanging dead leaves, and in dead leaves on ground (Rehn, 1945a).

Cariblatta imitans

Panama.—Among loose leaves on leaf mold in heavy jungle (Hebard, 1916a).

Cariblatta insularis

Jamaica.—One of the most frequently encountered orthopterous insects in bromeliads on trees (Hebard, 1916a, 1917; Rehn and Hebard, 1927).

Cariblatta jamaicensis

Jamaica.—In decaying herbage (Rehn and Hebard, 1927).

Cariblatta landalei

Jamaica.—All specimens taken from under drying bracts of banana blossoms (Rehn and Hebard, 1927).

Cariblatta lutea lutea

North Carolina.—Under pine straw on ground in woods (Brimley, 1908).

Southeastern U.S.—Under dead oak leaves; under dead needles in longleaf-pine woods; in wire grass; under refuse; beaten from undergrowth in pine and oak woods (Rehn and Hebard, 1916). In undergrowth of shortleaf-pine, longleaf-pine, and oak woods; in heavy scrub in damp spot of sand dune area; from high bushes, Ilex coriacea [= lucida] along inland swampy area (Hebard, 1916a). "The species is in large part terrestrial, being usually found among dead leaves and litter on the ground. Occasional specimens are, however, sometimes beaten from bushes. Individuals are decidedly active and are usually to be found in the greatest numbers in sandy situations" (Hebard, 1917).

Florida.—Throughout winter and spring they are frequent beneath leaves and other debris on ground, especially in dry, sandy locations (Blatchley, 1920). Friauf (1953) found this species under debris, fallen leaves, leaf mold, or decaying wood in these habitats: Dry, ruderal grassland (infrequent), scrub (frequent), sandhills (dominant), xeric hammock (infrequent), mesic hammock (dominant), pond margin (infrequent), longleaf-pine flatwoods (frequent), bayhead (occasional), low hammock (frequent), and alluvial hammock (occasional). In the shrub stratum in these habitats: Scrub (frequent), sandhills (dominant), and xeric hammock (infrequent). In herbaceous stratum in these habitats: Sandhills (dominant), mesic hammock (dominant), and black-pine flatwoods (infrequent). On bare soil or bare sand under vegetation in these habitats: Sandhills (dominant), pond margin (infrequent), longleaf-pine flatwoods (frequent), and slash-pine flatwoods (frequent) (Friauf, 1953).

Cariblatta lutea minima

Florida.—Series of specimens captured on Long Key under dead petioles of coconut palm on moist ground at edges of pools of brackish water. Specimens from Key West were in dry dead grass under boards (Rehn and Hebard, 1912). Nymphs frequent under bark on decaying pine logs in pine woods; occasional in leaf mold in heavy junglelike scrub (Rehn and Hebard, 1914). In water-soaked leaves in heavy red-mangrove swamp (Hebard, 1915). Under dead petioles of coconut palm on sandy soil in grapefruit grove (Hebard, 1916a). Numerous at bases of tufts of coarse grass growing just back of sea beach (Blatchley, 1920). Friauf (1953) found this species in leaf duff, leaf mold, debris, or decaying wood in these habitats: Dry, ruderal grassland (occasional), scrub (infrequent), sandhills (infrequent), mesic hammock (infrequent), pond margin (occasional), longleaf-pine flatwoods (occasional), and low hammock (infrequent). On bare soil or bare sand under vegetation in these habitats: Longleaf-pine flatwoods (occasional) and slash-pine flatwoods (occasional). Dominant in the spartina marsh habitat in the grass stratum and duff around clumps. Frequent in the saw-grass marsh habitat in the grass stratum and, during the dry season, in decaying vegetation on the marsh floor.

Cariblatta nebulicola

Jamaica.—Adults in dead leaf litter alongside the trail in dense forest of tree ferns, Podocarpus, Cyrilla, and other trees; the forest was bathed in fog much of the time (Rehn and Hebard, 1927).

Cariblatta reticulosa

Jamaica.—In leaves on leaf mold in hillside forest (Hebard, 1916a). Moderately numerous in leaf litter in mangrove swamp; in decaying herbage (Rehn and Hebard, 1927).

Cariblatta stenophrys

Puerto Rico.—Between the leaves and under the leaf sheaths of corn (Sein, 1923; Wolcott, 1936).

Cariblattaspp.

West Indies.—The tropical species of this genus inhabit heavy forest, living among the fallen leaves resting on the leaf mold, in epiphytic bromeliads, and in dead agaves (Hebard, 1916a; Rehn and Hebard, 1927).

Cariblattoides instigator

Cuba.—In siftings from under sea grapes, other shrubs, and low trees (Rehn and Hebard, 1927).

Cariblattoides suave

Puerto Rico.—On dry limestone hills (Rehn and Hebard, 1927).

Ceratinoptera picta

Trinidad.—Under bark of old cacao tree (Princis and Kevan, 1955).

Chorisoneura flavipennis

Costa Rica.—Under stones on borders of Surubres River (Rehn, 1906).

Chorisoneura formosella

Jamaica.—Swept from huckleberry trees (Vaccinium meridionale) (Rehn and Hebard, 1927).

Chorisoneura parishi

Panama.—From jungle undergrowth (Hebard, 1920).

Chorisoneura specilliger

Panama.—In grass (Hebard, 1920).

Chorisoneura texensis

Florida.—"The almost impenetrable jungle on Key Largo was examined, and in its depths the two specimens of this species were secured by beating the lower branches of gumbo limbo, other trees and the lower bushes and shrubs, among which latter are to be found such tropical forms as Ocotea catesbyana [= Nectandra coriacea] and Citharexylum villosum" (Rehn and Hebard, 1912). In nests of webworm and beaten from bushes of bayberry, Myrica cerifera, along edge of pine woods (Rehn and Hebard, 1916). Beneath dead leaves in oak woods and beaten from foliage of oak and bayberry (Blatchley, 1920). Infrequent in the tall shrub stratum of the xeric hammock habitat (Friauf, 1953).

Texas.—The great majority of specimens were beaten from foliage of bushes (Hebard, 1943a).

Southeastern and southern U.S.—In undergrowth in pine woods; beaten from shrubbery, from bayberry bushes, from lower branches of gumbo limbo and other trees, from lower bushes and shrubs in jungle, and from low oaks on hills. In Texas, beaten from tall weeds in opening in river-plain jungle scrub (Hebard, 1917).

Chorisoneura translucida

Panama.—In jungle vegetation, including vines covering low bushes (Hebard, 1920).

Chromatonotus infuscatus

Trinidad.—Males on low herbage under old cacao tree (Princis and Kevan, 1955).

Chromatonotus notatus

Trinidad.—Males in orchard on low herbage at night; females under refuse and in grass (Princis and Kevan, 1955).

Comptolampra liturata

Malaya.—Often found between dry foliage in the beakers of the epiphytic fern, Asplenium nidus, although the species lives mainly in bamboo bushes (Karny, 1924).

Cryptocercus punctulatus

North Carolina.—"They were never found except in parts of the logs [chestnut] where the decayed wood was soft, punky and wet" (Rehn and Hebard, 1910).

Oregon.—In fir logs where sap wood was soggy (Hebard, 1917).

Virginia.—In decaying chestnut and pine logs; taken six times in chestnut and once in pine (Hebard, 1917). In rotten logs in deep ravines of moist woods (Davis, 1926).

Appalachian Mountains, U.S.—In southern Virginia and eastern Tennessee, it is usually quite abundant in well-forested areas at elevations from 3,000 to 5,000 feet; "sometimes even a majority of the dead logs on a mountain side have roaches in them" (Cleveland et al., 1934). This cockroach not only lives in rotten, dead logs but also in sound logs that have been down only a few years. In Virginia it is found more often in chestnut and hemlock. "It occurs fairly often in oak, and has been found in pine, spruce, and arbor vitae.... There is little evidence that they ever leave the log and enter the ground" (Cleveland et al., 1934).

Cryptocercus relictus

Eastern Manchuria.—In great numbers under rotting fallen trees and in rotten dead wood (Bei-Bienko, 1950).

Cutilia soror

Marquesas Islands.—Males under stones and dead log (Hebard, 1933a).

Hawaii.—In soil about roots of pineapple (Illingworth, 1927). Often found about roots of grasses and weeds and other debris (Williams et al., 1931). Under stones and pineapple mulching paper (Fullaway and Krauss, 1945).

Wake Island.—Numerous, some from rotten logs. Found in bunch grass on Ocean Island (Bryan, 1926).

Cutiliaspp.

Australia.—Frequent woods where they leave shelter soon after sunset and run actively on ground or ascend shrubs and trees in quest of prey (Tepper, 1893).

Dendroblatta sobrina

Panama.—Colony on tree trunk; on surface of trunk of fallen tree (Hebard, 1920).

Diploptera punctata

Hawaii.—"Crowds of these insects in various stages of development sometimes gather in cypress trees, in suitable chinks, in old flowerhead sheaths of palms, etc., and even more or less openly on leafy twigs, in bunch grass, and the species is at times locally abundant behind the older leaf bases of sugar cane" (Williams et al., 1931). Williams also lists the following as food plants: Cryptomeria, algaroba, lime trees, ripening mangoes, papayas, and oranges. However, Bianchi (personal communication, 1954) doubted that any of the above are the main dietary, because the largest populations he had seen "were found in the fairly dry litter of Star Jasmine (Jasminum pubescens Willd.), well removed from any of the plants mentioned by Williams."

Raiatea, Society Islands.—Beaten out of bracken (Cheesman, 1927).

Uahuka, Marquesas Islands.—Under bark (Hebard, 1933a).

Dryadoblatta scotti

Trinidad.—Very common in water-filled, epiphytic bromeliads in the rain forest (see p. 31) (Princis and Kevan, 1955).

Ectobius africanus

Belgian Congo.—Females in forest margin and in forest undergrowth (Rehn, 1931).

Ectobius albicinctus

South France.—Females and young beneath stones (Blair, 1922).

Ectobius duskei

U.S.S.R.—In the steppe belt, it is a very characteristic member of feather-grass steppes, where it is found in associations of typically steppe vegetation, with feather grasses at the head (Stipa lessingiana and others), and on rocky slopes; it occurs frequently in cultivated fields of young crops and also in young geological strata in sections with virgin soil. The populations of this steppe cockroach average 6 to 8 individuals per square meter from the middle to the end of July. By the end of summer most individuals were observed at the bases of straw stacks with a canopy, having their south sides sheltered. This is the only species of Ectobius adapted to a purely steppe biocenose. (Bei-Bienko, 1950.)

Ectobius lapponicus

Southeastern Europe.—Numerous under stones on Trebovic (Burr, 1898).

U.S.S.R.—Found in wooded communities and peat bogs (in northern part of its range); males occur predominantly on herbaceous plants and bushes, but females hide under fallen leaves, moss, etc. (Bei-Bienko, 1950). It populated about 25 percent of the aspen trees in an experimental plot, feeding in galleries in the bark of young branches; there were 25 or more individuals per tree (Stark in Bei-Bienko, 1950).

Germany.—Abundant in woods; in pine woods in company with Stenobothrus vagans and Tettix kraussi. Numerous in low aspen bushes in forest. Numerous in deciduous and coniferous forests on trees and underbrush; under fallen leaves and moss; on oaks (Zacher, 1917). In foliage of young oak on top of mountain (Ramme, 1923).

Great Britain.—Under moss and dry leaves, among woodland undergrowth, and, generally, on vegetation close to the ground; occasional on bushes and trees (Lucas, 1920). Nymphs in heather in February and later; adults among rushes fringing pond in July (Lucas, 1925). Nymphs and males on rushy vegetation; unusually abundant on low herbage in dried-up swamp (Lucas, 1930).

Ectobius nicaeensis

France.—In dry woods, on bushes, and at the base of trees (Chopard, 1947).

Ectobius pallidus

Algeria.—Under stones; in moist places that are shaded and covered with plants (Lucas, 1849).

England.—Very abundant on sand dunes and among bracken in July (Buxton, 1914).

Germany.—In deciduous and coniferous forests; at edge of forest, from bare woods and bushes; numerous under leaves in oak woods and under moss (Zacher, 1917). In forest well lighted by the sun (Ramme, 1923).

Massachusetts.—Under loose lichens and bark on oak trees; under boxes, baskets, paper, etc., near houses; on Swiss chard (Flint, 1951). On roofs of houses, in shrubbery (Gurney, 1953). We have collected this species for several summers in a fairly dense, wooded area near dwellings, among fallen leaves and climbing on the erect stems and undersides of the leaves of periwinkle. OÖthecae were found on the ground under leaves and debris.

Ectobius panzeri

England.—Abundant on sandhills along shoreline among roots of grass (Burr, 1908). Under dead seaweed and other rubbish a few yards from shore on ground that would be washed by the sea (Lucas, 1896). Nymphs found among marram grass (Buxton, 1914). On sandhills near coast and covered with marram grass; often found on heather and low herbage; under old bark and rotten wood on posts; in decayed stump (Lucas, 1920). Swarming on Beta maritima and other plants in July (Lucas, 1920a). Very common in all stages in August, being frequently found under stones (Lucas, 1925). Common on sand dunes especially under stems of dead marram grass. Viable oÖthecae found buried in sand (Brown, 1952).

Germany.—In beech woods and in pine woods (Zacher, 1917).

Ectobius semenovi

Kazakhstan.—Along the shores of the Syr-daria it is found on and around living willows and on Populus euphratica; under loose bark of dying and dead trees (Bei-Bienko, 1950).

Ectobius sylvester

U.S.S.R.—In wooded steppe zones; probably only occurs in association with forests (Bei-Bienko, 1950).

Ectobius tadzhicus

Tadzhikistan.—Great numbers at the roots of Eleagnus shrubs on the banks of reservoirs and frequently under the bark of old trees (Bei-Bienko, 1950).

Ectobius vittiventer

South France.—One male beneath stone (Blair, 1922).

Ellipsidion affine

Australia.—From leaves, from scrub (Hebard, 1943). Collected in trees (Pope, 1953a).

Ellipsidion australe

Australia.—On eucalyptus leaves, on wattle, under bark (Hebard, 1943). Collected in trees (Pope, 1953a).

Ellipsidion bicolor

Australia.—In corn and from tree (Hebard, 1943).

Ellipsidion simulans

Australia.—From sugarcane (Hebard, 1943).

Ellipsidionspp.

Australia.—All stages are diurnal moving about the foliage of shrubs and small trees in bright sunlight on hottest summer days (Tepper, 1893).

Epilampra abdomen-nigrum

Trinidad.—In dried-up drain; among grass; in debris under old cacao tree; under old leaves (Princis and Kevan, 1955).

Puerto Rico.—Abundant in damp lowlands (SeÍn, 1923). Under dead leaves in wet malojillo meadow (Wolcott, 1936).

This species is amphibious (p. 31). Shelford (1907) suggested that immature stages of other species of the genus may be aquatic, which would place them in moist situations on the shores of rivers and other bodies of water.

Epilampra azteca

Panama.—Very scarce, under palm trees in decaying leaf mold and litter; one found under decaying bark of a log (Hebard, 1921a).

Epilampra mona

Mona Island, Puerto Rico.—One specimen under bark of dead tree (Ramos, 1946).

Epilampra tainana

Cuba.—Under dead leaves on stream bank (Rehn and Hebard, 1927).

Epilampra wheeleri

Puerto Rico.—In siftings from high-altitude primeval forest (Rehn and Hebard, 1927).

Epilampraspp.

Australia.—By day the insects live under bark, stones, logs, dead vegetable debris, or buried in loose dust or soil. After sunset females wander in grass or ascend low objects (Tepper, 1893).

Ergaula capensis

Uganda.—In open bush and short grass (Princis, 1955).

Eudromiella bicolorata

Panama.—Under rubbish on edge of jungle (Hebard, 1920).

Euphyllodromia liturifera

Colombia.—In brushwood (Princis, 1946).

Eurycotis biolleyi

Costa Rica.—Numbers of individuals were found in the large bromeliads of the temperate localities (Picado, 1913).

Eurycotis decipiens

Trinidad.—In old, rotten coconut stump (Princis and Kevan, 1955).

Eurycotis dimidiata

Cuba.—"This species was recorded from under stones in the fields ... by Gundlach" (Rehn and Hebard, 1927).

Eurycotis ferrum-equinum

Cuba.—Under stones in woods (Rehn and Hebard, 1927).

Eurycotis floridana

Florida.—Moderately common under bark of dead pine stumps and logs; at Key West it fairly swarmed under coquina boulders in the woods (Rehn and Hebard, 1905). Many specimens under palmetto leaves on ground (Caudell, 1905). In pine woods under dry bark of dead logs; on Long Key in dry fibers at the base of the heads of coconut palms; "at Key West, a large colony was discovered among boards lying on dry grass in a field, and several were captured upon turning over coquina boulders in the dense bush" (Rehn and Hebard, 1912). Particularly numerous in tree cavities and under bark along the edge of hammock areas (Hebard, 1915). Abundant between basal leaves of Tillandsia utriculata; beneath loose bark of logs and stumps; in and beneath decaying palmetto trunks and leaves; under rubbish (Blatchley, 1920). On ground in heavy tangle after dark; in decaying log of Sabal palmetto; in bromeliads; common under debris and bark in jungle; under signs on Pinus caribaea; in almost every sheltered outdoor place (Hebard, 1917). It moves about at night and hides under bark of logs and in other recesses during the day; where pines are present it almost invariably hides under bark of dead logs and stumps (Rehn and Hebard, 1914). Friauf (1953) found this species in leaf duff, leaf mold, or decaying wood in these habitats: Sandhills (infrequent), xeric hammock (dominant), mesic hammock (frequent), and low hammock (dominant); on tree trunks in sandhills habitat (infrequent) and mesic hammock (frequent); infrequent in saw-grass marsh habitat in the grass stratum and, during the dry season, in decaying vegetation on floor of marsh. Under the bark of logs and beneath logs in the woodpile habitat (Friauf, 1953).

Eurycotis galeoides

Cuba.—Under stones in deep woods (Rehn and Hebard, 1927).

Eurycotis kevani

Trinidad.—Under debris, trash, and vegetable refuse (Princis and Kevan, 1955).

Eurycotis opaca

Cuba.—In pine and palmetto region (Rehn and Hebard, 1927).

Euthlastoblatta abortiva

Texas.—Under dense tangle of bushy vegetation, palms, and vines near Rio Grande; in leaves and dry litter on ground; on dead petiole hanging from palm tree (Hebard, 1917). Under bark of dead hackberry; abundant in dead leaves, dry litter, and rats' (Neotoma sp.) nests in heavy scrub (Hebard, 1943a).

Graptoblatta notulata

Tahiti.—On foliage in sun or concealed among dead leaves that collect between the fronds of tree ferns (Cheesman, 1927).

Hawaii.—Quite active during the day, occurring on sugarcane, etc., in the wetter districts; it is also a household insect (Williams et al., 1931).

Hemiblabera brunneri

Puerto Rico.—Under bark of tamarind tree (Rehn and Hebard, 1927). Under the bark on a fence post (Wolcott, 1950).

Henicotyle antillarum

Dominica.—From rotting wood and wood soil (Rehn and Hebard, 1927).

Holocompsa metallica

Dominican Republic.—Along railroad through jungle and swamp (Rehn and Hebard, 1927).

Hololampra bivittata

Canary Islands.—Found in numbers among pine needles; nymphs were in the majority, adults rare (Burr, 1911).

Hololampra chavesi

Azores.—Very common in the hedges, particularly in brambles. Contrary to most species of this genus, which live on the ground under stones, this species is exclusively dendricolous and is only captured by beating the bushes on which it abounds (Chopard, 1932).

Hololampra maculata

Germany.—Abundant in deciduous forest in grass and under fallen leaves; in pine forests under lichens and between fallen needles; in edge of coniferous forest; under stones (Zacher, 1917).

Hololampra marginata

Macedonia.—Usually found crawling on the flowers and stems of giant thistles in May; common on thistles in June (Burr, 1923).

Hololamprasp.

Caucasus.—Numerous beneath dry leaves in a garden (Burr, 1913).

Hololeptoblattasp.

Seychelles.—Apparently only inhabits Pandanus between the leaf bases (Scott, 1910, 1912).

Homalopteryx laminata

St. Vincent.—In decaying leaves in forest (Rehn and Hebard, 1927).

Trinidad.—In forest debris and debris under old cacao trees; it is not uncommon under dry leaves; it feigns death when disturbed (Princis and Kevan, 1955).

Hormetica laevigata

Brazil.—From crown of palm between leaf bases (Hancock, 1926).

Ignabolivaria bilobata

U.S.S.R.—Under rocks and on the edges of woods in the lowlands in the north and in the mountains in the south (Bei-Bienko, 1950).

Ischnoptera deropeltiformis

North Carolina.—Under pine straw on ground in woods (Brimley, 1908).

Georgia.—Under dead oak leaves; under debris in garden; running on ground in pine and oak woods (Rehn and Hebard, 1916).

Indiana.—It is "a ground-frequenting, forest-loving insect, hiding beneath cover or about the edges of deep woodland, more frequently in damp places, and rarely taken beneath bark, signs, or at lights" (Blatchley, 1920).

Missouri.—Twenty to 30 males found resting on heads of wild oats on successive evenings (Rau, 1947).

Texas.—It preferred damp, open woodlands (Hebard, 1943a).

Eastern and southeastern U.S.—Under stone in heavy deciduous forest; under damp, dead leaves on edges of forests; under bark of pine log; in wire grass and sphagnum bordering stream thicket; in leaf mold and rubbish about pothole in pine woods, Pinus caribaea; under debris and leaf mold in hammock; under dead oak leaves in heavy deciduous forest (Hebard, 1917).

Florida.—"This species is distinctly geophilous and appears to prefer damp surroundings" (Rehn and Hebard, 1912). Under boards on very wet ground in everglades; in debris and leaf mold in heavy, junglelike areas of trees, bushes, and vines (Rehn and Hebard, 1914). Adults and numerous nymphs beneath weeds, grass, and other debris washed up on beach of Lake Okeechobee (Blatchley, 1920). Friauf (1953) found this species in leaf duff, leaf mold, and/or decaying wood on ground in these habitats: Dry, ruderal grassland (infrequent), scrub (infrequent), sandhills (occasional), xeric hammock (frequent), mesic hammock (dominant), shrubby, longleaf-pine flatwoods (infrequent), bayhead (dominant), and low hammock (dominant). On open bare soil or bare sand under vegetation in these habitats: Dry, ruderal grassland (infrequent), mesic hammock (dominant), moist, ruderal grassland (infrequent), pond margin (occasional), longleaf-pine flatwoods (infrequent), slash-pine flatwoods (infrequent), and low hammock (dominant). Infrequent in the herbaceous stratum of these habitats: Dry, ruderal grassland, moist, ruderal grassland, and longleaf-pine flatwoods. Infrequent in the shrub stratum of the dry, ruderal grassland habitat. (Friauf, 1953.)

Tennessee.—Taken in traps baited with cantaloupe in a parklike stand of oak, gum, hickory, and tulip trees in a creek bottom, and in a stand of oak on a dry ridge (Walker, 1957).

Ischnoptera panamae

Panama.—Under rubbish at edge of jungle and under drift on edge of coral-sand beach (Hebard, 1920).

Ischnoptera podoces

Jamaica.—In dead leaf litter along side trail through mountain forest (Rehn and Hebard, 1927).

Ischnoptera rufa rufa

Virgin Islands, St. Croix.—Common under rubbish and on shrubbery at night (Beatty, 1944).

Barbados.—Occasionally found in cane fields (Tucker, 1952).

West Indies.—In Puerto Rico, under stones in cultivated area, under debris on alkalie flat. In Jamaica, under dry petioles of coconut palm in grassy area; under logs, logwood on docks, and litter on limestone and near beach. In Panama, under drift on edge of coral-sand beach; under rubbish at edge of jungle (Hebard, 1916c).

Jamaica.—Under limbs and leaf litter in mangrove swamp (Rehn and Hebard, 1927).

Lamproblatta albipalpus

Panama.—Under drift on edge of coral-sand beach. Several under decayed banana stem (Hebard, 1920).

Lamproblatta meridionalis

Trinidad.—Under debris in forest and debris under old cacao trees (Princis and Kevan, 1955).

Latiblattella chichimeca

Costa Rica.—Very common in the bromeliads of all Costa Rica (Picado, 1913).

Latiblattella lucifrons

Arizona.—"Most commonly seen feeding on pollen and dead insects on the flower stalks of Yucca elata in June in the Santa Rita Mountains" (Ball et al., 1942).

Latiblattella rehni

Florida.—Widely distributed throughout pine woods (Pinus caribaea); under signs on Pinus clausa and Pinus caribaea (Hebard, 1917). Beneath bark of dead pine tree; beating Spanish moss; they seldom attempt flight when disturbed, but hide in crevices or drop to ground (Blatchley, 1920).

Latiblattella zapoteca

Costa Rica.—Under stones on borders of Surubres River (Rehn, 1906).

Leucophaea maderae

Barbados.—In cane fields (Tucker, 1952).

Dominica.—In vegetation of royal palms, guava, etc.; under loose bark and banana sheaths. In Jamaica, on logwood docks (Rehn and Hebard, 1927).

Litopeltis biolleyi

Costa Rica.—Under bark of tree in forest; in epiphytic bromeliads (Rehn, 1928).

Litopeltis bispinosa

Panama Canal Zone.—About 80 specimens from rotting banana stalks at bases of leaves; boring in decaying banana stem (Hebard, 1920).

Litopeltis deianira

Costa Rica.—In tree stump on edge of mountain forest; in dead wood on ground (Rehn, 1928).

Litopeltis musarum

Costa Rica.—Shaken from dead banana leaves. Footnote to specific name: "In relation to the liking of species of this genus for bananas (Musa) as shelter and possibly food" (Rehn, 1928).

Lobolampra subaptera

France.—Under stones and dead leaves, always rare (Chopard, 1947).

Loboptera decipiens

France.—All stages common beneath stones (Blair, 1922). Under stones and dead leaves (Chopard, 1947).

Maltese Islands.—Quite common in open country under stones (Valletta, 1955).

Dalmatia.—On seashores under rocks and seaweed cast up on shore (Bei-Bienko, 1950).

Loboptera thaxteri

Argentina.—Common in rubbish and leaf litter in small woodlot (Hebard, 1932).

Lobopterella dimidiatipes

Hawaii.—Abundant in wet districts, both in lowlands and to a considerable altitude in the forests, under trash, stones, boards, etc. (Williams et al., 1931). Often it is found with nymphs of Periplaneta australasiae (Fullaway and Krauss, 1945).

Lophoblatta arawaka

Trinidad.—On grass, maize, and cut sugarcane fodder; under vegetable and garden refuse; under old cacao (Princis and Kevan, 1955).

Macropanesthia rhinocerus

Australia.—Infrequently seen during dry season from March to October. "They burrow quite deeply, about two feet below the surface of the sandy soil in stands of cypress pine (Callitris sp.). They make a nest of dead leaves, grass roots, etc., frequently among the pine roots. The young nymphs rarely appear above ground, but following rain the adults burrow to the surface, especially at night.... This species is also found in the brigalow (Acacia harpophylla) scrub about 70 miles west of Rockhampton, Queensland, and on Fraser Island off the Coast of Queensland" (Henson in Day, 1950).

Megaloblatta blaberoides

Panama.—Under bark on tree (Hebard, 1920).

Ecuador.—Under a dense pile of dead leaves around base of tree (Campos R., 1926).

Megamareta verticalis

Australia.—In sugarcane (Hebard, 1943).

Methana canae

Australia.—Under loose bark on dead upright tree (Pope, 1953a).

Methana curvigera

Australia.—Under loose bark on trees and logs; many specimens on wattle trees where in strong sunlight they hid in curled-up leaves; oÖthecae attached to underside of loose bark and leaves (Pope, 1953a).

Methana marginalis

Australia.—Under loose bark of trees and logs (Pope, 1953a).

Moluchia (?) dahli

Chile.—Collected from lichens and mosses on tree trunks (Princis, 1952).

Muzoa madida

Costa Rica.—Under dead wood in dense second-growth forest; in thick mat of hanging dead vegetation in dense forest; under leaves in forest (Rehn, 1930).

Nauclidas nigra

St. Vincent.—Under rotten fruit (Rehn and Hebard, 1927).

Nelipophygus ramsdeni

Cuba.—Under rotten bark (Rehn and Hebard, 1927).

Neoblattella detersa

Jamaica.—Under dried leaves of coconut palm; in dry leaves under acacia on hillside; in debris on beach; under stones on coral rock; in leaf mold under dense brush on hillside; under bracts of banana blossoms (Rehn and Hebard, 1927).

Neoblattella dryas

Jamaica.—In bases of dead tree-fern fronds; numerous in bromeliads; nearly all collected specimens were taken in these plants (Rehn and Hebard, 1927).

Neoblattella eurydice

Jamaica.—Nearly all collected specimens taken in bromeliads (Rehn and Hebard, 1927).

Neoblattella grossbecki

Jamaica.—In epiphytic bromeliads and hollow bases of dead tree-fern fronds; nearly all collected specimens taken in bromeliads (Rehn and Hebard, 1927).

Neoblattella proserpina

Jamaica.—Under bark of huckleberry; nearly all collected specimens taken in bromeliads (Rehn and Hebard, 1927).

Neoblattella semota

Jamaica.—All specimens collected from under drying bracts of banana blossoms (Rehn and Hebard, 1927).

Nesomylacris cubensis

Cuba.—In dry region of palmettos and pines (Rehn and Hebard, 1927).

Nesomylacris relica

Jamaica.—Widely distributed from sea level to 5,700 feet elevation; in bromeliads in mountain forest; among dead leaves in heavy leaf mold under dense hillside scrub; under stones and in ground litter about banana trees; under bark of tree in dense ridge-type forest; in dead agave in scrub forest (Rehn and Hebard, 1927).

Nocticola bolivari

Ethiopia.—Always found under stones or cement blocks, but not necessarily deeply buried in the ground (Chopard, 1950b).

Nyctibora laevigata

Jamaica.—In cracks in dead stump of mimosa; in bromeliads (Rehn and Hebard, 1927).

Nyctibora lutzi

Puerto Rico.—Possibly to be found most often in rotten tree trunks in the highest mountains; found in rotten stump with termites, ants, and beetle grubs (Wolcott, 1950).

Nyctibora obscura

Trinidad.—Under pile of cornstalks (Princis and Kevan, 1955)

Nyctibora stygia

Haiti.—Under loose dead bark of mesquite tree, 52 specimens (Rehn and Hebard, 1927).

Oniscosomaspp.

Australia.—The females bury themselves in loose soil or dust (Tepper, 1893).

Opisthoplatia orientalis

Formosa.—On or in swampy ground or under rotten trees on the ground (Takahashi, 1924).

Panchlora antillarum

Dominican Republic.—In cultivated grounds, palms, fruits, etc. (Rehn and Hebard, 1927).

Panchlora nivea

Panama.—As Pycnosceloides aporus, in jungle under decaying banana stem in which were boring individuals of Litopeltis bispinosa (Hebard, 1920).

Texas.—Lives in foliage and in the green sheaths of plants (Hebard, 1943a).

Cuba.—On cane leaves; according to Gundlach this genus lives under the loose bark of trees (Rehn and Hebard, 1927).

Puerto Rico.—In rotting trunks of coconut palms (SeÍn, 1923). Most specimens have been collected from the very rotten interior of coconut palms (Wolcott, 1950).

Trinidad.—On corn; under old log; flies readily to lights (Princis and Kevan, 1955).

Panchlora sagax

Dominica.—In decaying stump in banana patch and in rotting wood. In Puerto Rico, in rotten coconut palm (Rehn and Hebard, 1927).

Panesthia australis

Australia.—In burrows under the thick bark of fallen and rotting trees (Shaw, 1914). In loose detritus, beneath clods of earth, and in fissures at foot of cliffs along the seashore beyond direct action of the waves (Tepper, 1893).

Panesthia laevicollis

Australia.—Under decayed logs in coastal scrub. It burrows into the soft part of the log (Froggatt, 1906).

Parcoblatta bolliana

North Carolina.—Under pine straw on ground in pine woods (Brimley, 1908).

Texas.—Under dry cow dung in pine woods (Hebard, 1917).

Nebraska.—Under pile of old boards (Hauke, 1949).

Parcoblatta caudelli

North Carolina.—From under the bark of dead trees (Rehn and Hebard, 1910).

Virginia.—At night on shrubbery. In South Carolina, under sign on tree (Hebard, 1917).

Tennessee.—In traps baited with cornmeal, cantaloupe, or fish in a stand of oak on dry ridge, and in abandoned rocky field on a south-facing slope (Walker, 1957).

Parcoblatta desertae

Texas.—From mountains, arid, and semi-arid regions; under small boulder on desert (Hebard, 1917). On ground in dry-creek bed through scrub oak, pine, and juniper forest (Hebard, 1943a).

Parcoblatta divisa

Eastern and southeastern U.S.—All specimens taken from under signs on red oaks and longleaf and shortleaf pines in Georgia and Virginia (Rehn and Hebard, 1916). Trapped in molasses-baited jar in oak forest in New Jersey; under signs on red and white oaks, sweet gum, and other deciduous trees; under signs on shortleaf and longleaf pines and pine stumps (Hebard, 1917). Widespread in southeastern U.S. in habitats as diverse as dry pine lands, oak scrub, moist hammocks in northern Florida, and deep, cool ravines along Apalachicola River (Hebard, 1943a).

Parcoblatta fulvescens

Eastern and southeastern U.S.—Trapped in molasses jars: in heavy, barrier-beach forest; in typical pine-barrens undergrowth; in pine barrens with heavy, grassy undergrowth; on border of pine barrens and on edge of swamp; in heavy deciduous forest; in heavy oak woods. Found under debris in dead, shortleaf-pine needles; under dead leaves on edge of oak and shortleaf-pine woods; under bark of pine log; among dead leaves under live oaks; under sign on Pinus caribaea (Hebard, 1917).

Georgia.—From under bark of pine log, among dead leaves under live oaks, and under leaves on edge of oak and shortleaf-pine woods (Rehn and Hebard, 1916).

Florida.—Very common among dead leaves, under logs, beneath loose bark, and wanders about at night in pinelands, hammock, turkey oak, and sand-scrub habitats (Hubbell and Goff, 1940). Beneath drift, cow dung, leaves, boards, bark of logs, and other debris, usually in open pine woods in sandy areas; frequent at the base of thistle leaves (Blatchley, 1920). Friauf (1953) found this species in leaf duff, debris, or decaying wood in these habitats: Scrub (dominant), sandhills (dominant), xeric hammock (dominant), mesic hammock, longleaf-pine flatwoods (infrequent), low hammock (infrequent), and alluvial hammock (infrequent). In the shrub stratum in these habitats: Scrub (dominant), sandhills (dominant), xeric hammock (dominant), and longleaf-pine flatwoods (infrequent). In the herbaceous stratum of the longleaf-pine flatwoods habitat, and under bark and beneath logs in the woodpile habitat.

Parcoblatta lata

Southeastern and southern U.S.—Under bark of pine logs and stumps; in sweet-gum logs and stumps; moderately numerous under bark of dead shortleaf pines; under bark of longleaf-pine stumps; under signs on red oak and longleaf pines; in dead oak. In Texas, under bark of pine stumps (Hebard, 1917).

North Carolina.—All stages under loose bark of dead pines, both prostrate and upright, and stumps. "It seems to prefer the space under the bark to be rather damp" (Brimley, 1908). Under bark of dead pine trees (Rehn and Hebard, 1910).

Florida.—Infrequent in leaf duff and decayed wood of low hammock habitat (Friauf, 1953).

Indiana.—Beneath rocks on sides and tops of high hills, in limestone glades where cedar abounds (Blatchley, 1920).

Missouri.—In leaf stratum of oak-hickory forest (Dowdy, 1951). Earlier, Dowdy (1947) reported finding numerous immature Pseudomopinae [presumably Parcoblatta sp.] in soil and leaf strata of oak-hickory forest.

Texas.—Captured in molasses-baited traps in low, wet, oak woods and in dry woodlot on hillside (Hebard, 1943a).

Parcoblatta pensylvanica

Eastern and southeastern U.S.—Trapped in molasses-baited jars; in oak and in chestnut forests, and on knoll with high deciduous trees. Found in oak and pine woods, under bark of decaying chestnut log and dead chestnut stump, and under signs on trees including oaks (Hebard, 1917).

North Carolina.—In all stages under loose bark of upright, dead pines, when the space under the bark was dry (Brimley, 1908).

Virginia, North and South Carolina.—Under signs on trees (white and red oaks); under bark of dead shortleaf-pine and sweet-gum logs and stumps (Rehn and Hebard, 1916).

Indiana.—Beneath bark of logs and stumps; empty oÖthecae common beneath loose bark of logs, especially shellbark hickory (Blatchley, 1920). Under loose bark on logs in January (Blatchley, 1895).

Illinois.—In pine forest associes, in black oak forest on sand, in oak-hickory forest on clay, and in climax forest; it evidently moved into the pine associes nightly, great numbers of oÖthecae were found under bark of pine logs, where, in October and November, hibernating nymphs were found (Strohecker, 1937). In nests of Vespula maculata (Balduf, 1936; McClure, 1936).

Missouri.—Usually in hollow trees, under loose bark, in woodpiles, and in cracks in rural buildings (Rau, 1940).

Michigan.—Common in oak-dune and beech-maple forests, under loose bark on dead trees and fallen logs, and under debris on forest floor (Hubbell, 1922). "A characteristic inhabitant of the low shrub-terrestrial and probably the terrestrial-hypogeic stratum." It occurred throughout the upland forests; groups were found established in and under logs 100 to 200 feet from the nearest forest (Cantrall, 1943).

Ontario.—Very abundant in rocky, sparsely-wooded country, where it occurred in rotten logs and under loose bark; on tree trunk at night on rocky island in lake (Walker, 1912).

Parcoblatta uhleriana

North Carolina.—Under pine straw on ground in woods (Brimley, 1908). Under bark of dead trees; 92 males attracted to lights (Rehn and Hebard, 1910).

Virginia.—Resting on woods foliage; at night on road (Rehn and Hebard, 1916).

Eastern and southeastern U.S.—Trapped in molasses-baited jars: in oak and pine woods, in heavy barrier-beach forest, in both scant and typical undergrowth on pine barrens, in heavy grassy undergrowth on pine barrens, on border of pine barrens, on edge of swamp, in heavy deciduous forest, in heavy oak woods, in upland oak and chestnut forest, in chestnut forest, in forested ravine, and on ridge with heavy oak, chestnut, and maple forest. Found under damp leaves on edge of forest, under bark of decayed chestnut log, inside decaying chestnut log with Cryptocercus punctulatus, under palmetto roots, under bark of pine stump, and in dry leaves under live oaks (Hebard, 1917).

Tennessee.—In traps baited with cornmeal or cantaloupe in maple-gum-oak forest in a mesic valley, and in a stand of oak on a dry ridge (Walker, 1957).

Indiana.—Beneath cover on slopes of high wooded hills. "This is essentially a forest-loving species; usually occurring beneath leaves and other debris on or along the borders of heavy hardwood timber." (Blatchley, 1920.)

Illinois.—In oak-hickory forest on clay and in climax forest (Strohecker, 1937).

Michigan.—In oak-dune woods (Hubbell, 1922). Restricted to woodlands, where it inhabited piles of moist dead leaves and rotten logs in oak-hickory forest (Cantrall, 1943).

U.S.A.—This species, P. uhleriana, and P. virginica were attracted at night to honeydew secreted by aphids on Pyrus sp. (Davis, 1918).

Parcoblatta virginica

New England.—Females under loose stones, boards, and other debris on ground; beneath loose bark (Morse, 1920).

North Carolina.—Under debris in dead shortleaf-pine needles (Rehn and Hebard, 1916).

Florida.—Infrequent in the shrub stratum of the scrub habitat. This was the only habitat of 25 studied in which this species was found (Friauf, 1953).

Eastern and southeastern U.S.—Trapped in molasses-baited jars: in pine and oak woods, in pine barrens, in pine woods with heavy grass undergrowth, in oak forest, in heavily forested ravine, on rocky slope with few deciduous trees, on knoll with high deciduous trees, in lofty chestnut forest, and in heavy low chestnut and oak forest on high ridge; under bark of decaying chestnut log and stump; under stones in chestnut forest; under bark of pine stumps (Hebard, 1917).

Indiana.—Frequents borders of open woods and fields; under debris, loose bark, and half-buried logs (Blatchley, 1920).

Illinois.—In black-oak forest on sand, in oak-hickory forest on clay, and in climax forest (Strohecker, 1937).

Michigan.—Common in oak-dune and beech-maple forests; under loose bark on dead trees and fallen logs and under debris on forest floor (Hubbell, 1922). Restricted to woodlands, where it inhabited piles of moist dead leaves and rotten logs in oak-hickory forest (Cantrall, 1943).

Texas.—Captured in molasses traps in moist woods of maple, oak, and pine with much undergrowth and a heavy layer of duff; in open, rather dry woodlot of Spanish oak and other trees; and in low wet woods of willow and oak along creek (Hebard, 1943a).

Parcoblatta zebra

Indiana.—Beneath log in cypress swamp (Blatchley, 1920).

Louisiana and Mississippi.—In decay cavity in sweet gum; under sign on shortleaf pine (Hebard, 1917).

Parcoblattaspp.

Alabama.—In the dry wall of a sweet-gum stump together with serropalpid and tenebrionid beetles (Snow, 1958).

Ohio.—OÖthecae under loose bark of fallen trees, where as many as 184 oÖthecae were found within a few feet of each other; others found under boards and in piles of firewood (Edmunds, 1952).

Pelmatosilpha coriacea

Puerto Rico.—Mona Island, under bark of dead trees and under guava leaves (Ramos, 1946). Under bark of Sideroxylon foetidissimum (Wolcott, 1941). Common along the coast and in mountains. "Very much at home" under the loose bark of Sideroxylon foetidissimum (Wolcott, 1950).

Pelmatosilpha kevani

Trinidad.—Under debris in bush (Princis and Kevan, 1955).

Pelmatosilpha purpurascens

Dominica.—In decaying logs in forest (Rehn and Hebard, 1927).

Periplaneta americana

Bermuda.—Among and under decaying debris, just above high-tide line (Verrill, 1902).

Johnson Island.—Nocturnal, coming out at night in great numbers about Tribulus blossoms. Under timbers on French Frigate Shoals (Bryan, 1926).

United States.—Alleyways and yards may be overrun during the summer; adults and hundreds of nymphs found in decaying maple trees along residential street (Gould and Deay, 1938, 1940). Around fumaroles where a railroad fill was burning internally (Davis, 1927). Common in palm trees along the gulf coast of Texas, where they often fly around street lights at night (Zimmern in Gould and Deay, 1940).

Periplaneta australasiae

Bermuda.—Very abundant under stones (Rehn, 1910).

Jamaica.—Under bark of dead tree and under bases of leaves of coconut palms (Rehn and Hebard, 1927).

Virgin Islands, St. Croix.—Common in sugarcane fields and in woodlands (Beatty, 1944).

Florida.—Juveniles under bark of dead logs of Pinus caribaea (Hebard, 1915). Frequently found under signs on trees near borders of towns; under bases of dead petioles of cabbage palmetto (Hebard, 1917). Beneath logs, burlap bags, and other cover in old orange orchards (Blatchley, 1920).

Marquesas Islands.—Under coconut fronds and grass (Hebard, 1935).

Nihoa Island.—Nymphs only, on Sida, Pritchardia, bunch grass, and about camp (Bryan, 1926).

Periplaneta brunnea

Georgia.—Under signs on oaks (Rehn and Hebard, 1916).

Florida.—Beneath bark of stump (Blatchley, 1920).

Periplaneta fuliginosa

Southeastern and southern U.S.—"This species is usually encountered out of doors, in or near towns. Over its range it is frequently found under signs on trees" (Hebard, 1917).

Phidon (?) dubius

Chile.—Collected from mosses and lichens on tree trunks (Princis, 1952).

Phoraspis spp.

Brazil and Guiana.—In grasslands, plantations of maize, sugarcane, and other plants on the borders of forests; the cockroaches were always found between the leaves which form the branches of the plants (Doumerc in Blanchard, 1837).

Phyllodromica brevipennis

Asia Minor and western Europe.—On ground among grasses; under moss and brushwood in mountain meadows (Bei-Bienko, 1950).

Phyllodromica graeca

U.S.S.R., western Georgia.—In pine forest mixed with deciduous trees (Bei-Bienko, 1950).

Phyllodromica irinae

U.S.S.R., Turan Lowland.—Along margins of "tugas" under half-fallen bushes of Salsola kali that overhang the ground (Bei-Bienko, 1950).

Phyllodromica maculata

Central Europe and western U.S.S.R.—On the edges of forests of the central-European type that are lighted by the sun; under fallen leaves; on bushes and conifers (Bei-Bienko, 1950).

Phyllodromica meglerei

U.S.S.R.—Among fallen leaves under bushes; on oak branches; under mown hay (Bei-Bienko, 1950).

Phyllodromica polita

Caucasus.—Under fallen leaves on slopes of mountains covered by forest or brushwood (Bei-Bienko, 1950).

Phyllodromica pygmaea

U.S.S.R.—In the sands of Un-dzhal-kum and Zhety-konur it is found in the dense turf of Aristida pennata (Bei-Bienko, 1950).

Phyllodromica tartara

Central Asia.—In lowlands and in mountains up to 2,500 meters; in fruit orchards under trap rings fastened to trees to combat lesser apple worm (Bei-Bienko, 1950).

Phyllodromica tartara nigrescens

Southern Uzbekistan.—Under bark of Juniperus sp., under stones and on flowers of Scorzonera acanthoclada (Bei-Bienko, 1950).

Platyzosteria castanea

Australia.—Under loose wood or bark (Shaw, 1914).

Platyzosteria novae seelandiae

New Zealand.—Swarms under loose dry bark and logs (Walker, 1904).

Plectoptera dominicae

Dominica.—On moss-covered lime trees. "The species of the genus Plectoptera are all foliage and flower frequenters, generally secured by beating low arborescent vegetation, or are attracted to light" (Rehn and Hebard, 1927).

Plectoptera dorsalis

Puerto Rico.—In caladium, grass, weeds, coffee, and bananas; in flowers of Ipomoea tiliasea (Rehn and Hebard, 1927). "... living in trees between leaves, or in 'butterfly-nests' of Tetralopha scabridella in leaves of Inga vera, or of Pilocrocis secernalis in the leaves of 'capÁ blanco' (Petitia domingensis) in the mountains. Along the coast they have been found under the bracts of cotton squares or bolls, and under the leaf-sheaths of sugar cane, in curled-up leaves of grapefruit, or in the dry flower clusters of 'espino rubial' (Zanthoxylum caribaeum)." These observations apply also to Plectoptera infulata and P. rhabdota (Wolcott, 1950).

Plectoptera floridana

Florida.—On fringe of tall bushes at edge of mangrove swamp (Rehn and Hebard, 1914). Rehn and Hebard (1927) stated that on the Keys it frequented dry scrubby vegetation, particularly Ilex cassine.

Plectoptera infulata

Puerto Rico.—See Wolcott's (1950) comments under Plectoptera dorsalis above.

Plectoptera lacerna

Cuba.—In grasses, sedges, etc., about a waterhole; on grass, pines and oak (Rehn and Hebard, 1927).

Plectoptera perscita

Dominica.—On moss-covered lime trees (Rehn and Hebard, 1927).

Plectoptera porcellana

Cuba.—Taken on flowers of "JÚcaro" (Gundlach in Rehn and Hebard, 1927).

Plectoptera pygmaea

Jamaica.—In relatively dense forest foliage; in shrubbery (Rehn and Hebard, 1927).

Plectoptera rhabdota

Puerto Rico.—In mixed vegetation; on grapefruit tree and guava (Psidium guajava); on bushes and shrubs (Rehn and Hebard, 1927). On coffee trees; on Spondias; on sugarcane; in caterpillar nests of Tetralopha scabridella on Inga vera; in old cotton bolls; on grapefruit (Wolcott, 1936). See also Wolcott's (1950) comments under Plectoptera dorsalis.

Plectoptera vermiculata

Cuba.—On pine in palmetto region (Rehn and Hebard, 1927).

Polyphaga aegyptiaca

Algeria.—Nymphal females under decaying leaves at the end of November (Lucas, 1849).

Transcaucasia.—In burrows in argillaceous cliffs along ravines. Females often covered by attached clay particles, an indication, according to Bei-Bienko, that this species is ecologically connected to compact clay soils or at least does not avoid them (Bei-Bienko, 1950).

See also the section on desert habitats (p. 29).

Polyphaga saussurei

South-central Asia.—Occupies compact clay soils; distributed in drier regions than P. aegyptiaca; frequently found near dwellings, in yards, stables, and houses (Bei-Bienko, 1950).

Polyzosteria limbata

Australia.—Common, usually "resting among the foliage or sunning itself on a fence or stumps, seldom or never hiding under bark or logs like most of the species" (Froggatt, 1906).

Poroblattaspp.

Tropical America.—"The species of Poroblatta apparently live as borers in stumps and logs in a manner similar to those of Cryptocercus Scudder in the United States" (Gurney, 1937).

Pseudomops septentrionalis

Texas.—In dead-brush pile; not scarce in heavy weeds, sunflowers, etc., in openings of river-plain jungle scrub (Hebard, 1917). It lives largely in herbage (Hebard, 1943a).

Pycnoscelus surinamensis

Florida.—Under stones and rubbish; very abundant under coquina boulders in woods at Key West (Rehn and Hebard, 1905). "This species is common under planks, stones, and other debris on the ground ... also found at Long Key in the dry fibres at the base of the petioles of a coconut palm" (Rehn and Hebard, 1912). At Musa Isle, found burrowing in sand (Hebard, 1915). In fallen leaves and decaying wood in xeric and mesic hammock habitats (Friauf, 1953).

Hawaii.—The soil swarmed with young of various stages during the summer (Illingworth, 1915). In soil about roots of pineapple under mulching paper; feeding on pineapple roots (Illingworth, 1927, 1929).

Fakarava, Tuamoto Archipelago.—Numerous among dead leaves in tree holes (Cheesman, 1927).

West Indies.—Under decayed stalks of sugarcane and in siftings from mangrove swamps, Cuba. Under manure, bases of leaves of coconut palm, litter, logs, and stones on coral rock and in bromeliads, Jamaica. Under wood, tiles, and boards in stable yards; immature individuals bored into the soil, Puerto Rico. (Rehn and Hebard, 1927.)

Barbados.—Frequents cane fields (Tucker, 1952).

Puerto Rico.—"Altho primarily a xerophytic species: collected among dry stones on Mona Island, under dry cow dung at BoquerÓn, and under boxes at GuÁnica, it is reasonably common in the more humid parts of Puerto Rico" (Wolcott, 1950).

Virgin Islands, St. Croix.—Common under rubbish; frequently seen feeding on chicken feces around chicken roosts (Beatty, 1944). By feeding on chicken feces it may become the vector of the chicken eyeworm, Oxyspirura mansoni, as described in the references cited on page 204.

Egypt.—Large numbers were found in moist soil at the site of a manure pile (Chakour, 1942).

Germany.—Under greenhouse conditions the depth to which P. surinamensis penetrated the soil was determined; 21 dug down to a depth of 8 to 10 cm., 3 dug down 10 to 12 cm., but only one dug 13 cm. below the surface. Often the tubes in the soil ended in a chamber which the cockroach might not leave for several days; nymphs molted in such chambers and females bore their young there (Roeser, 1940).

Rhytidometopum dissimile

Trinidad.—Male on low herbage in orchard at night; under sacking; on Hibiscus at night (Princis and Kevan, 1955).

Riatia orientis

Trinidad.—Numerous specimens of both sexes at night on roadside Hibiscus rosa-sinensis or low herbage in orchard (Princis and Kevan, 1955).

Simblerastes jamaicanus

Jamaica.—Numerous in fragmentary debris of an abandoned termite nest on ground in the dry Liguanea Plain; a specimen was also taken under a stone in a field of short grass (Rehn and Hebard, 1927).

Styphon bakeri

Costa Rica.—Among humus and rubble in crevices and large cavities in rocks of the Tertiary limestone rim and the metamorphosed and igneous rocks of the interior of the islands (Baker in Rehn, 1930).

Supella supellectilium

Virgin Islands, St. Croix.—Under rubbish heaps; in sugarcane straw (Beatty, 1944).

Africa.—"A cosmotropical species which occurs both out of doors and as a household pest in many warmer parts of the world. It is apparently endemic to non-forested areas in much of Africa north of the Equator." (Kevan and Chopard, 1954.)

Symploce flagellata

Puerto Rico.—Under low trees on hillside and dead leaves in thicket of sea grape (Hebard, 1916c).

Symploce hospes

Hawaii.—Under stones and rubbish (Illingworth, 1915).

Virgin Islands, St. Croix.—Under rubbish and on shrubbery at night (Beatty, 1944).

Symploce jamaicana

Jamaica.—In dead leaves under acacia and other shrubs in desert tract; under log and rubbish in open on limestone sand near beach (Hebard, 1916c). Very common in short dry grass in roadside gutter at night, often clustered together; under beach trash in stony wash of Hope River (Rehn and Hebard, 1927).

Symploce ruficollis

Virgin Islands, St. Croix.—Under rubbish and on shrubbery at night (Beatty, 1944).

Puerto Rico.—In siftings from sea-grape thicket on sandy soil (Rehn and Hebard, 1927). Often living under leaf-sheaths of sugarcane (Wolcott, 1950).

Tartaroblatta karatavica

Asia, Kara-tau Mountains.—Many hundreds of individuals found only under stones on moist earth and not where ground seemed dry; found on very stony slopes with sparse vegetation, often with undergrowth present (Bei-Bienko, 1950).

In this category we include all man-made structures, whether inhabited by man or not, that may become infested with cockroaches. A nonexhaustive list of such structures would include dwellings, restaurants, mess halls, barracks, groceries, markets, bakeries, dairies, drug stores, department stores, hotels, hospitals, warehouses, mills, factories, packing houses, animal houses, breweries, incinerators, privies, sewers, sewage treatment plants, ships, aircraft, etc. Although dwellings are only one of the many kinds of structures that are colonized by cockroaches, the several species that have adopted this mode of life are generally referred to as domiciliary cockroaches. This term is adequate only if we remember that these cockroaches are not restricted to domiciles but are pests in other structures as well.

Associations between man and certain species of cockroaches possibly started as casually as the short-lived association that Beebe (1953) observed when he discovered three cockroaches in the newly built couch of an orang-utan. Obviously, when man came down from the trees, his fellow travelers found his cave dwellings and other abodes particularly favorable habitats. From such primitive beginnings, domiciliary cockroaches have spread into every kind of structure that man has since devised. We predict that when man develops a suitable vehicle, cockroaches will someday accompany him into space. Yet despite the apparent predilection of certain species of cockroaches for man, man is only incidental to these associations. Only the shelter and food that man unwittingly provides for these unwelcome guests attract cockroaches to him; man's physical presence is unnecessary.

Most, if not all, of the common domiciliary cockroaches apparently originated in the Tropics or sub-Tropics from whence they have spread, through normal commercial channels, into most of the inhabited world. At least eight domiciliary cockroaches originated in Africa (Rehn, 1945): Blatta orientalis, Blattella germanica, Leucophaea maderae, Nauphoeta cinerea, Oxyhaloa buprestoides, Periplaneta americana, P. australasiae, and Supella supellectilium; and, perhaps, Periplaneta brunnea as well; Neostylopyga rhombifolia was probably of Indo-Malayan origin; Pycnoscelus surinamensis was of oriental origin; and Leurolestes pallidus was endemic in the West Indies (Rehn, 1945). Princis (1954a) rejected Africa as the original home of Blatta orientalis and advanced reasons for placing its origin in Central Asia.

Several domiciliary species have become well established in temperate zones and some even in the Arctic. Bei-Bienko (1950) listed the following 10 species as sinanthropes in the Palearctic zone: Blatta lateralis, B. orientalis, Blattella germanica, Leucophaea maderae, Periplaneta americana, P. australasiae, Polyphaga saussurei, Pycnoscelus surinamensis, and Supella supellectilium. In the warmer parts of the temperate regions, as in their native Tropics, certain domiciliary species breed outdoors as well as indoors. In the less temperate extensions of their ranges most domiciliary species are nearly always found indoors. In regions with low winter temperatures these cockroaches do not survive in unheated structures; but in heated buildings Blattella germanica, for example, has been able to withstand the rigorous climate of Alaska, where it has caused severe infestations (Chamberlin, 1949).

The limiting factors that determine whether man-made structures will provide suitable habitats for cockroaches are favorable temperature and availability of water and food. The range of temperatures that man provides for his own comfort and protection fosters the rapid increase of cockroach populations indoors. Gunn (1934, 1935) has demonstrated that the preferred temperature range (zone of indifference) of Blatta orientalis is 20-29° C. The upper limit of the preferred temperature of Blattella germanica and Periplaneta americana is 33° C. (Gunn, 1935). The lower limits of temperature tolerance were not sharply defined in Gunn's work. However, less than optimum temperatures, if they last for only short periods, are not necessarily lethal. The 24-hour mortality for P. americana that had been held for one hour at 0° C. was only 2±2 percent (Knipling and Sullivan, 1957). Gunn (1934) observed that Blatta orientalis would not settle at temperatures above 33° C. and would react violently against higher temperatures (e.g., 39° C.) by running away; thus the thermotactic behavior of cockroaches might be presumed to bring them into favorable environments within structures. Thermal death points have been determined for the above three species by Gunn and Notley (1936).

It is common knowledge among those who rear cockroaches experimentally that, unless the water content of the food is high, fluid water is essential in the insects' dietary. Ten species of domiciliary cockroaches have been shown to be unable to survive as long on dry food alone as they could on food and water at 36-40 percent relative humidity (Willis and Lewis, 1957). Blatta orientalis, when in a state of normal water balance, usually spent more time in the drier part of a humidity gradient; but desiccated insects tended to become hygropositive (Gunn and Cosway, 1938). We presume that other domiciliary species behave similarly. If water is available nearby, it may be presumed that partially desiccated cockroaches could locate a source through the mediation of a humidity sense. Hygroreceptors have been demonstrated on the antennae of Blattella germanica (Roth and Willis, 1952a) and suggested for Blatta orientalis (Gunn and Cosway, 1938).

Drinking water is available to cockroaches in the traps of sinks, wash basins, tubs, and toilet bowls; in flush tanks; as condensation on cold pipes, flush tanks, and windows; around leaking pipes and faucets; as spillage; in miscellaneous water-filled containers, such as pet drinking dishes, aquaria, vases; empty beverage bottles; and drainage from ice boxes. Soft, juicy fruits and vegetables can provide both moisture and food. There seems to be a tendency for certain species (Blatta orientalis and Blattella germanica) to become established in the more humid parts of structures, such as basements, around sinks, and in bathrooms. Whether this is a reaction to a preferred humidity or merely a fortuitous aggregation near sources of drinking water and food has never been clearly demonstrated. The rather widespread dissemination of these species into zones of low as well as high humidity suggests that detailed studies of the microclimatic conditions of structural microhabitats will be needed before meaningful conclusions can be drawn about the stratification of cockroaches within structures according to species.

In nearly all structures infested by cockroaches, food of some kind is available, either in the structure itself or nearby. This may be the food stored by man for his own use or the use of kept animals; it may be crumbs, food spillage, garbage, or excreta; glues and pastes on cartons, boxes, stamps, envelopes, labels, and wall paper; sizing on cloth and book covers; various dried animal and plant products; dead insects; living plants; etc. In fact, it is almost impossible, despite good housekeeping, to keep any structure used by man free of all food suitable for cockroaches.

That the requisite temperature, water, and food are provided, more or less adequately, by a variety of structures is attested by the innumerable infestations of cockroaches that develop when control measures are relaxed. Within structures the accessibility of certain harborages to cockroaches probably depends on the habits of the species and to some extent on their size. Similar types of harborages in different structures may be used by the same species, although there seems to be some overlapping by different species into the same kinds of daytime shelters. The comparative ecology of domiciliary cockroaches has not been thoroughly investigated, so any interpretation of observational data is necessarily speculative and inconclusive at this time. Our discussion on pages 324 to 343 is also pertinent to this section.

LAND-BASED STRUCTURES

Dwellings provide a variety of microhabitats that are acceptable to cockroaches. It has been stated that old houses, or houses that have many cracks and crevices, or have basement kitchens that are not kept clean and in good repair are particularly liable to invasion by cockroaches (Laing, 1946; British Museum [Nat. Hist.], 1951). Although this statement is undoubtedly true, it has been our personal experience, as well as the experience of others, that new, clean, and well-planned houses and apartments are also easily and sometimes quickly invaded by cockroaches. Mallis (1954) has cited the following places that are frequently infested by cockroaches in homes. In the kitchen, cockroaches are found in and around sinks, in cupboards above and below sinks, under tables and chairs, in stoves, around breadboards, in utility cabinets, in kitchen closets, under linoleum, behind, under, and inside refrigerators and iceboxes. In living rooms cockroaches are found in furniture, studio couches, sewing machines, closets, and bookshelves; behind picture frames, pennants, calendars, and other wall ornaments. In bathrooms cockroaches are found in and behind utility cabinets and toilets; they may be found in wicker clothes hampers, in brooms and mops, and in door hinges. Ordinarily, cockroaches are not found in bedrooms unless they are abundant elsewhere in the dwelling. Additional harborages are cited under specific cockroaches in the list below.

In markets DeLong (1948) found the German cockroach in bags of potatoes and onions, in crates of citrus fruits, in pads and shredded papers in banana boxes, and in cases of bottled beverages. The insects were attracted by coffee and crawled into the folds of coffee bags. They were found in cartons of canned goods; in bread and baked goods; in cartons of packaged cookies, cakes, and crackers. Packaged cereals were attractive, and cockroaches were sometimes found in packages of cigarettes. The insects occurred in scales (by the hundreds) and in cash registers. Rather heavy infestations were found under stainless-steel cappings that covered wooden arms on the fish cleaning stand. The insects were numerous in display cases where they were warm and sheltered. They were also found behind mirrors above produce racks, in electrical switch boxes and conduits, and in telephone boxes, as well as generally in cracks and behind loose moldings or loose wall boards. Enclosed boxlike tables were frequently heavily infested.

In restaurants cockroaches may be found in the following places: Crevices in wood, plaster, concrete, and metal; in the bar; in the kitchen and in the associated equipment; in cupboards, lavatories, and garbage storage areas; and on the undersides of chairs and tables (Mallis, 1954).

In drug stores Frings (1948) found cockroaches behind the mirror and between the sink and the cooler. Thousands were found in hollow ornamental shelf edging. The hollow bases of malted-milk dispensers and drink mixers were cockroach havens.

In a hospital Frings (1948) found cockroaches in decorative trim around doorways, by the thousands in wicker laundry baskets, and in incubators for premature babies. In military hospitals we have seen cockroaches (Blattella germanica) in kitchens and dining halls in the usual hiding places mentioned above and on the undersides of stainless-steel serving tables.

In department stores cockroaches have been found in food departments, beauty salons, rest rooms, dressing rooms, linen departments, and stationery departments (Anonymous, 1952). The infestation in the linen department was traced to clean towels which, when returned from the laundry, contained at least 500 cockroaches per bundle. The insects were carried into the rest rooms and beauty salon when the towels were distributed.

The microhabitats of cockroaches in privies and sewers have not been studied. These habitats are particularly important in view of the demonstrated migrations of cockroaches from sewers into dwellings and the possible dissemination of pathogenic microorganisms from feces to food. The reader is referred to our 1957(a) paper for a summary of the known information on cockroach dispersal from sewers.

COCKROACHES ASSOCIATED WITH LAND-BASED STRUCTURES

Most of the cockroaches listed below are either known domiciliary species or they have been found one or more times in houses or other man-made structures. The known structural pests breed within the building. Certain other species, which have been observed only infrequently in structures and are not known to breed there, may possibly be incipient pests; these latter species may attain future economic importance if they establish breeding colonies within a structure. A few species have undoubtedly wandered indoors by accident. It is difficult to decide whether a particular species was an accidental invader or whether it was attracted indoors in response to some stimulus. Only additional information will provide the desired answers.

Aglaopteryx ypsilon

Trinidad.—Male found indoors (Princis and Kevan, 1955).

Allacta similis

Hawaii.—Found only indoors at Nauhi. Otherwise this is apparently an outdoor species (Swezey and Williams, 1932).

Blaberus craniifer

Cuba.—Household pest (Deschapelles, 1939). Particularly abundant in houses in Santiago and Havana (Rehn and Hebard, 1927).

Florida.—Under boards in woodshed (Rehn and Hebard, 1912, 1914).

Blaberus discoidalis

Ecuador.—In eating places (Campos R., 1926).

Hispaniola.—In houses (Rehn and Hebard, 1927).

New Jersey.—In greenhouse (Weiss, 1917).

Puerto Rico.—In homes (SeÍn, 1923). In fruit stores (Wolcott, 1950).

Blatta lateralis

Central Asia.—Household pest, often found in homes with clay floors (Bei-Bienko, 1950).

Turkmen S.S.R.—Males and females occurred in dwellings (Vlasov, 1929).

Blatta orientalis

This species is a cosmopolitan domiciliary pest (Hebard, 1917; Rehn, 1945). It is reported to occur particularly in basements and crawl spaces under basementless houses (Mallis, 1954). In damp basements where food is available large colonies are not unusual, but it also may infest offices and apartments several floors off the ground (Gould and Deay, 1940). The number encountered on upper floors is seldom large, but the frequency of occurrence may reach 30 percent of the observations (Spear et al., in Shuyler, 1956). In supermarkets this species hides during the day inside concrete blocks or cracks in the foundation, under furniture, or behind cartons; it is conspicuous on the floors of the markets at night (De Long, 1948). In Great Britain the kitchen is preferred by this pest (and by Blattella germanica); they shelter beneath steam radiators and gas stoves, behind hot-water pipes, underneath furniture and floor coverings, sinks and baths; basements and underground kitchens are especially likely to be infested (Laing, 1946; British Museum [Nat. Hist.], 1951). Goodliffe (1958) noted that B. orientalis may travel long distances to find food.

Blattella germanica

This species is a cosmopolitan domiciliary pest (Hebard, 1917; Rehn, 1945). It is one of the commonest insects in homes and restaurants (Gould and Deay, 1940). It is found in kitchens, larders, bathrooms, furnace rooms, and storage rooms of bakeries, breweries, hospitals, barracks, as well as dwellings, where, during daylight, it hides behind cupboards, furniture, hanging pictures, panels and skirting boards, in cracks around drains, water pipes, electric wires, and hot-water and steam heating units (Wille, 1920). The German cockroach may be found in cracks around baseboards, pipes, conduits, sinks, and drawers; behind cabinets; inside switch boxes and refrigerators; on under surfaces of tables, chairs, and shelves; between stacks of stored goods, and in almost every place that is not readily observed (Kruse, 1948). We have also seen this species packed in electric-clock cases and loud-speaker baffles, in cash registers, and clinging to the undersurface of stainless-steel steam tables. The infestation of markets by this species has been described above. Very narrow cracks provide refuges for the German cockroach. Wille (1920) found first-instar nymphs in cracks 0.5 mm. wide and adult males and females without oÖthecae in cracks 1.6 mm. wide.

Shuyler (1956) has observed extensions into relatively new structural habitats by Blattella germanica in the north-central area of the United States. A few German cockroaches are now being encountered in living rooms, bedrooms, clothes closets, bedroom furniture, lobbies, entrance halls, checkrooms, nonfood storerooms, nonfood warehouses, and coin-vending machine repair shops. In these situations this species is behaving much like the brown-banded cockroach, Supella supellectilium.

Blattella schubotzi

Cameroon.—Five specimens in a house (Princis, 1955).

Blattella vaga

U.S.A.—Although this is mostly an outdoor species, during dry seasons it may temporarily enter houses in great numbers, occasionally breeding indoors in Arizona (Flock, 1941a). Two adults were collected indoors in Texas (Riherd, 1953).

Chromatonotus notatus

Trinidad.—A male was found indoors (Princis and Kevan, 1955).

Cutilia soror

Hawaii.—Almost as common in houses as Neostylopyga rhombifolia (Hebard, 1922).

Ectobius duskei

U.S.S.R.—Frequently occurs in living apartments in farming localities as an accidental inhabitant (Bei-Bienko, 1950).

Ectobius pallidus

Massachusetts.—A summertime pest in houses along coast (Gurney, 1953; E. R. Willis, personal observation). Generally an outdoor species.

Epilampra abdomen-nigrum

Trinidad.—Male, indoors (Princis and Kevan, 1955). An outdoor species generally.

Ergaula capensis

Cameroon.—A male taken in a house (Princis, 1955).

Eublaberus posticus

Trinidad.—Indoors, feeding on bat feces (Princis and Kevan, 1955).

Eurycotis floridana

Florida.—Occasionally found in homes (Creighton, 1954; Roth and Willis, 1954a).

Euthyrrhapha pacifica

Hawaii.—Found outdoors and indoors where it breeds in neglected cupboards and in rubbish (Fullaway and Krauss, 1945).

Holocompsa azteca

Mexico.—Household pest (Ball et al., 1942).

Holocompsa cyanea

Costa Rica.—One specimen in house (Rehn, 1906).

Holocompsa nitidula

Apparently domiciliary in American Tropics (Hebard, 1917). In houses under chests, etc., Cuba (Gundlach, 1890-1891); Puerto Rico (Gundlach, 1887). In folds of burlap bag, Florida (Rehn and Hebard, 1914).

Ischnoptera rufa occidentalis

Panama.—Thrives about human habitations under litter, though not domiciliary (Hebard, 1920).

Ischnoptera rufa rufa

Jamaica.—In hotel. "While hardly a domiciliary form it would seem to frequent environments where man has considerably disturbed natural conditions, as under debris, docks, under logs and stones in cultivated areas" (Rehn and Hebard, 1927).

Leucophaea maderae

West Indies.—In habitations, warehouses, and other structures; "At times it is a very abundant and serious pest" (Rehn, 1945). In Puerto Rico it was also found in fruit stores, markets, and inns (SeÍn, 1923; Wolcott, 1950).

Reported as a domiciliary pest in Madeira (Heer, 1864); Windward Islands (Marshall, 1878); Tropics and sub-Tropics (Rehn, 1937); Philippine Islands (Uichanco, 1953); New York City (Anonymous, 1953; Gurney, 1953); Trinidad (Princis and Kevan, 1955). This species is also established in coastal Brazil, Central America, all the Greater Antilles, several other tropical islands, and tropical Africa, where it probably originated (Rehn, 1945).

Leurolestes circumvagans

Hispaniola, Grenada.—Largely domiciliary (Rehn and Hebard, 1927).

Leurolestes pallidus

Cuba.—All over island, in houses, under lockers, etc. (Rehn, 1945; Gundlach, 1890-1891).

Florida.—Rehn and Hebard (1914).

This species has been recorded from various islands in the West Indies, from Mexico, Guatemala, and Brazil (Rehn, 1945).

Methana marginalis

Australia.—Reported entering houses (Pope, 1953a).

Nauphoeta cinerea

Australia.—In hospital (Mackerras and Mackerras, 1948). In dwellings, grain stores, and fowl-feeding pens (Pope, 1953).

Sudan.—Domiciliary in huts of the Shilluk natives; fairly widely distributed in eastern Africa (Rehn, 1945).

Hawaii.—In feed rooms of poultry plants (Illingworth, 1942).

Florida.—Established in feed mills around Tampa (Gresham, 1952; Gurney, 1953).

The wide distribution of this species from East Africa, where it originated, to the Orient and the New World was undoubtedly mediated by shipping (Rehn, 1945).

Neoblattellasp.

Puerto Rico.—Observed [as Blatta caraibea, Rehn and Hebard (1927)] in houses (Gundlach, 1887).

Neostylopyga rhombifolia

Domiciliary in Indo-Malaya and New World Tropics (Rehn, 1945); Philippine Islands (Uichanco, 1953); and Hawaii (Hebard, 1922).

Oxyhaloa buprestoides

Presumably to some extent domiciliary, as it evidently spread from Africa to the New World via slave ships (Rehn and Hebard, 1927; Rehn, 1945).

Panchlora nivea

Colombia.—A male and a female taken in a dwelling (Princis, 1946). This is primarily an outdoor species which is frequently taken indoors as an adventive on bananas (see p. 150 for references).

Parcoblatta fulvescens

Florida.—Males found in laboratory and dormitory buildings, ostensibly attracted by lights (Friauf, 1953).

Parcoblatta lata

Connecticut.—Domiciliary pest (Moore, 1957). Generally an outdoor species.

Parcoblatta notha

Arizona.—It may occasionally be a nuisance in houses (Ball et al., 1942).

Parcoblatta pensylvanica

U.S.A.—Country houses often badly infested, Indiana (Blatchley, 1920). Frequently taken in houses in wooded areas, Michigan (Hubbell, 1922). Infestation by males, females, and nymphs on fourth floor of building, South Dakota (Severin, 1952). Houses in wooded areas infested by nymphs and occasionally by adults (Gould and Deay, 1940).

Canada.—Pest in summer cottages in Ontario (Walker, 1912).

Periplaneta americana

This is a cosmopolitan domiciliary pest (Hebard, 1917; Rehn, 1945). It is common in restaurants, grocery stores, bakeries, and where food is prepared or stored; it was trapped regularly in the basement and upper floors of store buildings, and it was also found in all heated parts of an old meat-packing plant (Gould and Deay, 1938). P. americana was numerous in latrines in Iran (Bei-Bienko, 1950) and in privies in Texas (Dow, 1955) and Georgia (Haines and Palmer, 1955). Large numbers of this species also occur in sewers adjacent to human habitations (Roth and Willis, 1957a).

Periplaneta australasiae

Generally domiciliary, but also occurs outdoors in the West Indies (Rehn and Hebard, 1927). It is a very abundant domiciliary pest in tropical Africa and tropical America (Rehn, 1945); Ecuador (Campos R., 1926); Puerto Rico (Sein, 1923); Philippine Islands (Uichanco, 1953); Australia (Pope, 1953).

Also occurs as a pest in greenhouses in Pennsylvania (Thilow and Riley, 1891); France (Giard, 1900); Italy (Boettger, 1930); Great Britain (Laing, 1946; British Museum [Nat. Hist.], 1951).

Periplaneta brunnea

Circumtropical domiciliary pest which is apparently more nearly peculiar to the Tropics and adjacent regions than P. americana (Hebard, 1917). This species has been trapped in significant numbers in privies and dwellings in Georgia (Haines and Palmer, 1955). It is the species of Periplaneta present in homes in San Antonio, Texas (Mallis, 1954).

Periplaneta fuliginosa

U.S.A.—Frequently encountered out of doors, but has been reported common after dark about a hotel in Alabama and was captured in a house in Louisiana; it was also extremely abundant on the wharves at night in Jacksonville, Florida (Hebard, 1917). As a domiciliary pest it was, next to Blattella germanica, the most common cockroach inside homes in southwest Georgia, where it was also the most common cockroach in privies (Haines and Palmer, 1955). This species has become a very common domiciliary pest in Texas (Eads, personal communication, 1955). It infested a greenhouse for five years in Indiana (Gould and Deay, 1940).

Periplaneta ignota

Australia.—It occurs in dwellings occasionally (Pope, 1953).

Phaetalia pallida

Colombia.—Three specimens from three dwellings (Princis, 1946).

Trinidad.—Male indoors; male and female at light (Princis and Kevan, 1955).

Plectoptera dorsalis

Puerto Rico.—According to Gundlach (1887) it enters houses at night attracted by light (Rehn and Hebard, 1927).

Polyphaga aegyptiaca

Iraq.—Common in houses (Weber, 1954).

Caucasus.—Winged male in kitchen (Burr, 1913).

U.S.S.R.—Listed as a sinanthrope (Bei-Bienko, 1950).

Polyphaga saussurei

South-central Asia.—One of the commonest domiciliary species (Bei-Bienko, 1950).

Pseudophoraspis nebulosa

East Indies.—This species is sometimes difficult indoors (Karny, 1925).

Pycnoscelus surinamensis

A household pest in the East Indies (Karny, 1925); Philippine Islands (Uichanco, 1953); Tanganyika (Smith, 1955); Trinidad, eight records indoors (Princis and Kevan, 1955). It is also a greenhouse pest (Hebard, 1917; Zappe, 1918; Doucette and Smith, 1926; Saupe, 1928; Roeser, 1940). Common in or around chicken batteries and yards in Hawaii (Schwabe, 1949).

Supella supellectilium

Domiciliary wherever distributed (Rehn, 1945), this species is especially difficult to control because of its apparently nonselective dispersal throughout dwellings. For example, Mallis (1954) observed in Texas that it was widely distributed throughout the apartment and was probably the most common cockroach seen in the bedroom; its favorite harborages were beneath and behind corner braces on kitchen chairs, underneath tables, behind pictures and other objects on walls, and in shower stalls; its oÖthecae were commonly fastened on walls and ceilings throughout the house. Gould and Deay (1940) reported that this species prefers high locations, such as shelves in closets, behind pictures, and picture molding; oÖthecae were found about kitchen sink, desks, tables, and other furniture, and even in bedding. Hafez and Afifi (1956) stated that the adult wanders in nearly all rooms of the house and only visits the kitchen when searching for food; it hides in cupboards, pantries, closets, bookshelves, drawers, and behind picture frames; the nymphs normally hide in the corners of drawers, behind frames, and in similar situations.

Symploce bicolor

Puerto Rico.—In houses, Sardinera Beach, Mona Island (Ramos, 1946).

Symploce hospes

North American Tropics.—Domiciliary, but not exclusively so, and apparently widely distributed (Hebard, 1917). In Florida, as Ischnoptera rufescens, found in a greasy cupboard (Rehn and Hebard, 1914).

Hawaii.—Illingworth (1915).

SHIPS

Sailing ships have long been notorious for their unwelcome hordes of cockroaches, and it was by ship that at least 11 domiciliary species migrated from their centers of origin to other parts of the world (Rehn, 1945). Over 40 nondomiciliary species have been carried by ship from the American Tropics to other parts of the world in cargoes of bananas (p. 146). In addition to these, other adventive cockroaches appear from time to time in ports to which they have been carried by ships. Yet by far the most numerous cockroaches on shipboard are the breeding populations of a few common domiciliary pests. Except in the most rigorously disinsectized ships, this commerce in cockroaches has continued to the present day.

Cockroaches undoubtedly infested the first ships that sailed the Mediterranean; of these we have no records. The earliest recognizable record of cockroaches on shipboard is Moffett's (1634) statement that when Drake captured the ship Philip, he found it overrun with cockroaches [Blattarum alatarum]. Bligh (1792) described disinfesting H.M.S. Bounty with boiling water to kill cockroaches. Chamisso (1829) reported that he had seen ships casks, in which rice or grain had been stored, that were found to be filled with Blattella germanica when opened. During a voyage from England to Van Diemen's Land, Lewis (1836) was greatly annoyed by hundreds of cockroaches flying about his cabin at night; the most numerous resembled Periplaneta americana and another was similar to Ectobius lapponicus. This latter was undoubtedly B. germanica, which is the only ship-infesting species that resembles the feral E. lapponicus. Lewis continued, P. americana "were in immense profusion, and had communication with every part of the ship, between the timbers or skin. The ravages they committed on everything edible were very extensive; not a biscuit but was more or less polluted by them, and amongst the cargo 300 cases of cheeses, which had holes in them to prevent their sweating, were considerably damaged, some of them being half devoured and not one without some marks of their residence."

Kingsley (1870), Kellogg (1908), Gates (1912), Heiser (1936), and Bronson (1943) have all reported that cockroaches were so numerous on ships that they gnawed the skin and nails of the men on board. These are all independent observations of what may well have been a common occurrence on ships. We have discussed in detail the subject of cockroaches biting man in our 1957(a) paper.

Mosely (1892) reported, "At the time that England was left the ship [H.M.S. Challenger] seemed nearly free of animals, other than men, dogs, and livestock required for food. The first cockroaches apparently came on board at St. Vincent, Cape Verdes.... Cockroaches soon became plentiful on board, and showed themselves whenever the ship was in a warm climate.

"At one period of the voyage, a number of these insects established themselves in my cabin, and devoured parts of my boots, nibbling off all the margins of leather projecting beyond the seams on the upper leathers."

Sir Edmund Freemantle (1904) recalled some of his experiences in the British Navy. "Cockroaches in the tropics were also terrible scourges. One saw little of them in fine, dry weather, but in damp, wet weather they seemed to come from every hidden corner ... our remedy in the 'Spartan' was to make the boys catch them—on pain of being caned.... One brig, the 'Lily,' was so overrun by cockroaches that the officers' clothes smelt of them."

The quotation from Sonan (1924) on page 348 describes similar conditions in the Japanese Navy.

On modern cargo ships cockroaches are reported to be extremely numerous in the galley, the crew's quarters, and sometimes in the holds; they dwell in hot, humid environments such as the casing around steam pipes (Monro, 1951). Williams (1931) reported that Blattella germanica was the most important cockroach pest on ships seen at New York. Although often numerous in the holds, the cockroaches as a rule congregated in living quarters. They were also frequently found between tarpaulins covering the hatches. It was not unusual to kill 20,000 to 50,000 in the forecastle, and more than 20,000 have been taken from a single stateroom. Simanton (1946) inspected the S.S. William Kieth when it berthed at San Francisco from a 10-month voyage to the South Pacific. The holds were infested with thousands of Periplaneta americana, but in the crew's quarters, mess halls, and storerooms B. germanica predominated. After insecticidal treatment about 2,000 P. americana were seen in each hold and as many as 24 B. germanica in each cabin. Richardson (1947) reported that in Army transports inspected between 1943 and 1946 at New York, P. americana was found in the galleys and messes, and occasionally heavy infestations were found deep in the holds; B. germanica was found in the galleys and messes; Blatta orientalis was found only in the hold.

Additional references indicating the presence of Blattella germanica on ships may be found in the account of its parasite Ripidius pectinicornis (p. 232). Although Rice (1925) and Williams (1931) cite B. germanica as the most numerous cockroach on ships, Brooke (1920) stated that the great majority of ship cockroaches were Periplaneta americana. In addition to citing cockroach infestations on ships, the following authors reported various methods for disinfecting ships: Canalis (1916), Pryor (1918), Brooke (1920), Rice (1925), Williams (1931), Simanton (1946), Richardson (1947), and Anonymous (1951, 1954).

COCKROACHES ASSOCIATED WITH SHIPS

In the following list we include some previously unpublished data on cockroaches that were recovered from ships at the Miami, Fla., Quarantine Station for the periods November 1945 through May 1946; May, June, August, and September 1950; and 17 July 1957 (Porter, personal communication, 1958). These data were lumped, without breakdown to species, under the entry Orthoptera in Porter (1958).

Certain of the species listed below occur only accidentally on shipboard and will probably never establish breeding colonies on ships or become pests on shipboard or elsewhere; some were merely passengers between one land-based colony and another. Others, the truly domiciliary pests, are as likely to be pests on shipboard as they are in land-based structures.

Blaberus discoidalis

Hispaniola.—On board ship (Rehn and Hebard, 1927).

Blatta orientalis

U.S.A.—At Port of New York (Richardson, 1947).

Blattella germanica

At sea?—In ships casks (Chamisso, 1829).

U.S.A.—Port of New York (Williams, 1931; Richardson, 1947). San Francisco (Simanton, 1946). At Miami, 7,852 live specimens recovered from ships (Porter, personal communication, 1958). Most numerous species on ships (Rice, 1925).

Epilampra maya

At sea.—One male and one female found dead on S.S. Tenadores (Hebard, 1917).

Epilamprasp.

Florida.—One dead specimen, Miami (Porter, personal communication, 1958).

Ischnopterasp.

Florida.—Five live and one dead specimen, Miami (Porter, personal communication, 1958).

Latiblattellasp.

At sea.—One female alive in hold of S.S. Tenadores (Hebard, 1917).

Leucophaea maderae

Brought from West Africa to West Indies and Brazil by slave ships (Rehn, 1945).

Nauphoeta cinerea

Widely disseminated by sailing ships (Rehn, 1945).

Neoblattella fratercula

At sea.—Two females found dead on S.S. Tenadores (Hebard, 1917).

Neoblattella fraterna

At sea.—One male found dead in hold of S.S. Tenadores (Hebard, 1917).

Neoblattella nahua

At sea.—One female dead in hold of S.S. Tenadores (Hebard, 1917).

Neoblattellasp.

Florida.—Five dead specimens recovered from ships, Miami (Porter, personal communication, 1958).

Neostylopyga rhombifolia

Widely distributed by sailing ships (Rehn, 1945).

England.—Captured on a sugar vessel from Java (Lucas, 1920).

Nyctibora noctivaga

At sea.—One male and one female nymph found dead on S. S. Tenadores (Hebard, 1917).

Nyctiborasp.

Florida.—Two dead specimens recovered from ships at Miami (Porter, personal communication, 1958).

Oxyhaloa buprestoides

Spread from Africa to New World by ships (Rehn and Hebard, 1927; Rehn, 1945).

Panchlora nivea

At sea.—One female dead in hold of S.S. Tenadores (Hebard, 1917).

Florida.—Fifteen dead specimens taken from ships, Miami (Porter, personal communication, 1958).

From the numerous records of this species as an adventive taken on bananas (p. 150), it may be presumed to be a frequent traveler on banana boats.

Periplaneta americana

At sea.—Lewis (1836). Hebard (1933a) stated that this is "often a serious pest on the smaller ships sailing the South Seas."

U.S.A.—San Francisco (Simanton, 1946). Port of New York (Richardson, 1947). At Miami, 62 live and 123 dead specimens (Porter, personal communication, 1958).

Periplaneta australasiae

Migrated from West Africa to America in slave ships (Rehn, 1945).

Pycnoscelus surinamensis

Probably in part reached the New World by way of Africa in slave ships (Rehn, 1945).

Supella supellectilium

Reached America from West Africa by slave ship (Rehn, 1945).

Xestoblatta festae

At sea.—One female found dead in hold of S.S. Tenadores (Hebard, 1917).

AIRCRAFT

Michel (1935) stated that the development of air transportation brought the same insect dispersal problems that exist in land and water transportation; in addition, the problem of cockroach infestation had become a very serious one, quite aside from the hygienic point of view, because it had been discovered that these insects seek out the wings of airplanes, where they subsisted on the glue and dope used in airplane construction. However, Dethier (1945) found no cockroaches in dismantled or wrecked wing and tail structures of metal aircraft in central Africa. In fact, all-metal aircraft would seem to provide little in the way of food or water for stowaway cockroaches.

Laird (1951, 1952, 1956a) found living specimens of Blattella germanica, Periplaneta americana, and Periplaneta australasiae in baggage compartments and/or kitchens in aircraft. Other species which have been recovered from undisclosed spaces in aircraft are listed below. Some of the cockroaches that were reported as dead may not have died from exposure during flight but may have been killed by insecticide applied by inspecting personnel at the airports.

COCKROACHES ASSOCIATED WITH AIRCRAFT

In the following list we include some previously unpublished data on cockroaches that were recovered from aircraft in Miami, Fla., International Airport from 1 July 1956 through 30 June 1957 (Porter, personal communication, 1958). These data were lumped under the entry Orthoptera without breakdown to species in Porter (1958).

Species reported by Hughes (1949), and cited below as from southern United States, were recovered from aircraft that arrived at Brownsville, Fort Worth, Miami, New Orleans, and San Juan. There was no way of linking a specific record with any particular city.

The comments we made above about species that are infrequently encountered on ships apply with equal validity to similar species found on aircraft.

Anaplectasp.

U.S.A.—One live and 15 dead specimens recovered from 16 aircraft at Miami (Denning et al., 1947).

Blatta orientalis

U.S.A.—Six live and four dead specimens recovered from six aircraft at Miami (Denning et al., 1947). Southern U.S. (Hughes, 1949).

Blattella germanica

Hawaii.—Williams (1946a).

Khartoum.—Whitfield (1940).

New Zealand.—Laird (1951, 1952, 1956a).

U.S.A.—At Miami 193 live and 184 dead specimens were recovered from 141 aircraft (Denning et al., 1947). Recovered at airports in southern U.S. (Hughes, 1949). Recovered at Miami, 51 live and 24 dead specimens (Porter, personal communication, 1958). Exposed experimentally in jet aircraft (Sullivan et al., 1958).

Blattellasp.

Khartoum.—Whitfield (1940).

Southern U.S.—Hughes (1949).

Cariblattassp.

U.S.A.—One live and three dead specimens recovered from three aircraft at Miami (Denning et al., 1947). Southern U.S. (Hughes, 1949).

Epilamprasp.

U.S.A.—One dead specimen, Miami (Denning et al., 1947). Southern U.S. (Hughes, 1949).

Eublaberus posticus

Southern U.S.—Hughes (1949).

Ischnoptera rufa rufa

U.S.A.—Two dead specimens recovered from two aircraft at Miami (Denning et al., 1947).

Ischnopterasp.

U.S.A.—Two live and one dead specimen recovered from three aircraft, Miami (Denning et al., 1947). Southern U.S. (Hughes, 1949). One dead specimen, Miami (Porter, personal communication, 1958).

Leucophaea maderae

U.S.A.—Three dead specimens recovered from three aircraft at Miami (Denning et al., 1947). Southern U.S. (Hughes, 1949).

Nauphoeta cinerea

U.S.A.—One live and one dead specimen, Miami (Denning et al., 1947).

Neoblattellasp.

U.S.A.—One dead specimen recovered at Miami (Denning et al., 1947).

Panchlora nivea

U.S.A.—Two dead specimens recovered from two aircraft, Miami (Denning et al., 1947).

Periplaneta americana

U.S.A.—Five live and three dead specimens recovered from seven aircraft, Miami (Denning et al., 1947). Southern U.S. (Hughes, 1949). Three live and five dead, Miami (Porter, personal communication, 1958). Experimentally exposed in jet aircraft, U.S. (Sullivan et al., 1958).

New Zealand.—Laird (1951, 1952).

Periplaneta australasiae

U.S.A.—Two dead specimens recovered from two aircraft, Miami (Welch, 1939). Five live and three dead from five aircraft, Miami (Denning et al., 1947). Southern U.S. (Hughes, 1949). Three live and two dead, Miami (Porter, personal communication, 1958).

New Zealand.—Laird (1952).

Periplanetaspp.

U.S.A.—One live and three dead specimens from four aircraft, Miami (Denning et al., 1947). Southern U.S. (Hughes, 1949). One live and six dead, Miami (Porter, personal communication, 1958).

Pycnoscelus surinamensis

U.S.A.—Two live and three dead specimens from five aircraft, Miami (Denning et al., 1947). Southern U.S. (Hughes, 1949).

New Zealand.—Laird (1956a).

Supella supellectilium

U.S.A.—Two live and one dead specimen from one aircraft, Miami (Denning et al., 1947). Southern U.S. (Hughes, 1949).

Khartoum.—Whitfield (1940).

Supellasp.

U.S.A.—Two live specimens from two aircraft, Miami (Welch, 1939). Southern U.S. (Hughes, 1949).

Symplocesp.

U.S.A.—Two live and one dead specimen from three aircraft, Miami (Denning et al., 1947). Southern U.S. (Hughes, 1949).

Unidentified cockroaches

Anglo-Egyptian Sudan.—At Khartoum (Whitfield, 1940).

Brazil.—From flying boats, 62 specimens; from land planes, 45 specimens (Carneiro de MendonÇa and Cerqueira, 1947).

Central Africa.—Dethier (1945).

Kenya.—At Kisumu (Symes in Whitfield, 1940).

New Zealand.—Laird (1956a).

U.S.A.—Four live and 14 dead specimens from 4 aircraft, Miami (Welch, 1939). One live and 15 dead specimens (adults?) from 16 aircraft; 8 oÖthecae from 7 planes; 147 live and 83 dead nymphs from 108 planes, Miami (Denning et al., 1947). Southern U.S. (Hughes, 1949). Five live and 7 dead specimens, Miami (Porter, personal communication, 1958).

---