Many cells in which diseased brood was found were used again for brood and for storage of honey or pollen after the bees had removed all visible diseased material from them. The bees seemed able to clean out all infectious material efficiently. Only in the heavily diseased colonies did any of the new brood reared in such cleaned-out cells become infected. It appears that honey in most cases would not become contaminated when stored in such cells. This does not preclude the eventual spread of the disease organisms to stored honey or to other healthy larvae by the bees engaged in the cleaning process.

Brood from resistant and nonresistant colonies was found to be equally susceptible to American foulbrood infection when reared under controlled conditions in the center of heavily infected brood nests of diseased colonies. This fact suggests that colony resistance to American foulbrood depends on behavior factors, like the housecleaning abilities of the bees, rather than on actual physiological resistance in the larvae. No evidence of physiological resistance to the disease in honeybee larvae was found, other than the decrease in susceptibility associated with their age. Brood susceptibility was found to be greatest during the first day of larval life. It decreased thereafter to the extent that larvae more than 2 days and 5 hours old did not become infected. This period corresponds closely to the period of mass feeding of the larvae. The failure of some brood to develop disease when reared under infectious conditions may be the result of the short time in larval life when inoculation must occur to produce infection, and to the type of feeding at that time.

We obtained conclusive information that substantiated these observations as the result of the development of new method for the controlled inoculation of individual honeybee larvae of known age without causing general contamination of the colony. This method also has been applied as an additional test for determining a high level of resistance in colonies. It consists of adding, by means of a special microsyringe, known numbers of spores of Bacillus larvae in a small drop of water suspension to the food of the larvae within 1 day after they hatch from the egg. The exact locations and disposition of the inoculated larvae are determined for further observations on the development of disease with the aid of the same cell-locating frame used in previous studies: The comb of inoculated larvae is then returned to the colony for feeding and sealing of the brood.

Disease was produced by this method of inoculation in the brood of resistant colonies as readily as in that from susceptible ones. All diseased brood was removed by the bees of the most highly resistant colonies before any symptoms of the disease were apparent. It was found that Bacillus larvae in the vegetative rod stage is noninfective. In the resistant colonies the bees allowed no diseased brood to remain long enough to permit the organisms again to reach the infective spore stage, but in a susceptible colony infected larvae were allowed to remain long enough for spore formation to occur.

The conclusions to be drawn from the results of these various investigations are that resistance to American foulbrood in the honeybee colonies of resistant strains consists in the ability of the worker bees to detect and remove diseased brood before the causative organism, Bacillus larvae, reaches the infectious spore stage in the diseased larvae. That is what occurs in the so-called "negative colonies" considered most highly resistant in the apiary investigations of the development of resistance to American foulbrood. This is a purely behavior character that is developed and transmitted to succeeding generations through intensive breeding.

Up until the past few years, beekeepers in the United States attached relatively little importance to the disease of adult bees known as Nosema disease, caused by a protozoan parasite, Nosema apis. Like sacbrood, it has been accepted as a troublesome disease of minor importance, to be more or less controlled by good beekeeping practices. Recent investigations have shown that the disease is much more widespread through most of the country than beekeepers had supposed. Furthermore, we learned that Nosema disease is responsible for many cases of winter dysentery, weakening of overwintered colonies, and spring dwindling. There is considerable evidence to show that many queen bees are superseded and lost because they are infected. Package-bee colonies tend to show the effect of Nosema disease on brood rearing more than normal colonies and they are more likely to lose their queens because of the disease.

Recent laboratory studies with bees in cages disclosed that temperature materially influences the development of Nosema disease. The causative organism, Nosema apis, develops in the intestinal tract most rapidly at temperatures near 88 F., or slightly below the normal temperature of the active brood nest. Development was checked at about 93 or 94 F., which is near normal brood-rearing temperatures, and was completely inhibited above 98 F. Development also was checked by temperatures below the optimum and was completely inhibited at temperatures below 51 ' F. These temperature relations may explain the slower development of the disease during the broodless period of the winter, increased development of infection with the start of brood rearing in the later winter, and decreased development with the rapid increase of brood rearing and active bee flights in the spring. Thus, increased flight activity, the production of young bees, and higher brood-nest temperatures accompanying the active season may help either to eliminate infection or control its further spread among the bees of strong colonies. In further experiments, infected bees held in cages at 99 F. for 14 days appeared to have recovered completely from the infection. Whether some form of heat treatment for Nosema disease would be practical is not yet known.

The term "paralysis" has long been used to designate various little-understood disorders of adult honeybees that are characterized by trembling, sprawled legs and wings, occasional partial hairlessness, and, in some cases, a black and shiny appearance. The last two symptoms are not always as common as the other symptoms. Often a heavy death rate results, and the disease may affect whole colonies.

Recently in laboratory experiments in which material prepared from bees affected with so-called paralysis was fed to bees in cages under controlled conditions, we established that at least one type of paralysis is caused by a filtrable virus. A virus is an ultramicroscopic organism so small it passes through porcelain filters that hold back ordinary bacteria.

The discovery that a virus causes paralysis of adult bees may lead to finding a preventive or cure. With this more accurate knowledge of symptoms, beekeepers can judge better when it is necessary to requeen or otherwise manipulate infected colonies. The maintenance of strong colonies headed by young, vigorous queens is about the best preventive known at present. Beekeepers should watch primarily for signs of trembling, with sprawled legs and wings, as the most reliable symptoms of the paralysis disease in the hive.

THE AUTHOR

A. P. Sturtevant is an apiculturist in the Bureau of Entomology and Plant Quarantine. After graduation from Clark University, and 2 years of graduate study at Massachusetts Institute of Technology, he started his first research work on bee diseases as assistant in comparative pathology at the Massachusetts Agricultural Experiment Station in 1915. Since 1926 he has been in charge of the Intermountain States Bee Culture Laboratory of the Division of Bee Culture, Bureau of Entomology and Plant Quarantine at Laramie, Wyo., where his principal research interest has been with the various investigations concerning American foulbrood, and American-foulbrood-resistance studies in particular. He received a doctorate from George Washington University in 1923.