26 BULLETIN 780, U. S. DEPARTMENT OF AGRICULTURE. 
of the other colonies of the apiary (Table IV, colonies numbered 36, 
50, 66, 68, and 82). 
The results obtained indicate, therefore, that by inserting combs 
from Nosema-infected colonies, as was done in experiment No. 2, the 
infection is not transmitted appreciably. An explanation for this is 
easily seen from results recorded throughout the present paper. 
Further experiments on the point are summarized in Table XXVI. 
EXPERIMENT NO. 3. 
In this experiment 7 colonies free from Nosema infection were 
inoculated by feeding them sirup to which Nosema apis had been 
added. The bees from which the parasites were obtained for this 
experiment were from various sources (p. 12). They had been dead 
and drying in the laboratory at room temperature for at least three 
months. All of the 7 colonies received the first inoculation feeding 
on October 8. On each succeeding day for four days the feeding 
was repeated. Each of the inoculated colonies of the experiment 
was examined from time to time, but no Nosema-infected bees were 
found. The final examination in connection with this experiment 
was made on October 28. Out of 70 bees examined from the 7 
colonies only one Nosema-infected bee was found. The infection in 
this instance probably did not result from the inoculations. 
In this experiment it is shown that Nosema apis drying in the 
abdomen of bees at room temperature for three months does not 
produce infection when fed to healthy bees. This result suggested 
the interesting fact that the parasite of the bee resisted drying for a 
comparatively short time only (see other experiments, p. 40). 
EXPERIMENT NO. 4. 
In experiment No. 4, four of the colonies used in experiment No. 3 
were inoculated on October 29, 1912, with Nosema apis taken from 
infected bees recently killed. Nine days after the inoculation sam- 
ples of bees were examined from each of the four colonies inoculated. 
Nosema infection was found in nearly all of the bees examined. 
Two weeks after inoculation 50 bees were examined from each of 
the 4 colonies. All of the 200 bees were found to be infected. At 
the end of three weeks a similar condition prevailed. On Decem- 
ber 16, 48 days after inoculation, all of the 4 inoculated colonies 
were alive. A large number of bees were now found on the bottom 
board of the hive. By this time the colonies had become very much 
weakened. The bees were uneasy, the cluster being easily disturbed. 
During the following week 1 of the colonies died out completely. 
The remaining 3 were chloroformed. Another colony inoculated in 
November gave like results, and died in January, 1913. 
Each of the 5 colonies of the experiment were four-framed nuclei. 
As the inoculations were made late in the autumn there were no 
