[VIII. FUNGI AND YEASTS]
By far the greatest number of fungi known to be associated with cockroaches belong to the Laboulbeniaceae, genus Herpomyces, the species of which are restricted to parasitizing cockroaches (Thaxter, 1908). Most species are hyaline, small and inconspicuous (Thaxter, 1931) and are usually, but not exclusively, found on the insects' antennae. Species of Herpomyces are highly, but not completely, host specific (Richards and Smith, 1954). While attached to the host, these fungi appear like minute dark-colored, yellow, or white (e.g., H. arietinus) bristles or bushy hairs (pl. [27], A).
Fig. 1.—Diagram illustrating the relationship between a mature plant of Herpomyces stylopygae and the integument of Blatta orientalis. (Reproduced from Richards and Smith [1956], through the courtesy of Dr. A. G. Richards.)
Richards and Smith (1955, 1955a) have studied the life history of Herpomyces stylopygae on the oriental cockroach. The plants grow only on living cockroaches, and the infection is disseminated by contact. The mature plants are found mostly on the antennae (pl. [27], B), either on setae or on hard or soft cuticle. Spores are ejected from perithecia singly or in groups of 2 to 4 spores, although groups as large as 12 spores have been found. The presence of single, paired, or multiple spore groups on the surface of the host was correlated with the presence of single, paired, or multiple plants on infected cockroaches. Development from spore to mature perithecia takes about two weeks. The plant obtains nutriment from the host by means of a tubular haustorium that extends through the cockroach's cuticle and expands into a large bulb in the underlying epidermal cells (fig. 1). Infections on nymphs are lost when the nymph moults, but infections on adults persist throughout life. However, nymphs which have lost the fungus upon moulting are readily reinfected. Collart's (1947) statement that nymphs are never infected with Herpomyces is not true.
Richards and Smith (1956) concluded that there is no evidence of pathogenicity in Herpomyces infections because heavily infected cockroaches appear fully active in laboratory colonies; they can run at the same speed as uninfected cockroaches; they reproduce normally and do not appear to die prematurely. These workers stated that the infections cause a dermatitis which is neither pathogenic nor debilitant. So far as we know there are no comparative data on longevity and reproductive performance of fungus-infected versus normal cockroaches. However, Gunn and Cosway (1938) have shown that the presence of these fungi (identified as Stigmatomyces sp.; see p. 138) on the antennae seemed to interfere with the humidity reactions of Blatta orientalis. Although Richards and Smith (1956) admit that humidity receptors and other sense organs on the antennae may be destroyed by the fungus, they state that "insects possess such a large number of sensilla that the result may well be more distressing to the sensory physiologist than to the insect." Yet it seems to us that the loss of sense organs from fungal infection and concomitant shortening of the antennae (pl. [27], A) might be considerably more of a handicap to free-living cockroaches than those in laboratory colonies.
Bode (1936) studied the flora of Periplaneta americana and cultured Aspergillaceae and Mucorinae from the insect's body surface and intestinal contents; he also found nonsporulating yeasts in P. americana. To prevent fungal growth on oöthecae of P. americana, Griffiths and Tauber (1942a) autoclaved their rearing containers and dipped the oöthecae in 70-percent alcohol for 10 seconds.
Mercier (1906) isolated and cultured a pathogenic yeastlike parasite which had invaded the fat body and blood of Blatta orientalis. The abdomens of the infected insects became swollen, distended, and soft. McShan (unpublished MS., 1953) consistently isolated Saccharomycetes from the feces of Periplaneta americana.