12 BULLETIN 96, U. S. DEPARTMENT OF AGRICULTURE. 



of spring the mean of the cluster temperature also raised. Finally 

 when the days had considerably lengthened and were relatively 

 warm, the amplitude of the cluster variations increased, as is shown 

 in the schematic curve (fig. 7). When the summer season for the 

 bees began, accompanied by the beginning of incubation, the tem- 

 perature of the center of the cluster rose to 34° C. (93.2° F.) or 35° 

 C. (95° F.) and continued practically at this level. For the winter, 

 then, it might be said in a general way that the temperature prevail- 

 ing for several days is in a measure an index of the temperature of 

 the cluster. 



TEMPERATURE BELOW FRAMES IN RELATION TO OUTSIDE AIR. 



The thermometer /, situated below the bottom of the frames and 

 cluster, as is shown in the general views of the apparatus (figs. 1 and 

 2), registered the temperature of the air at the bottom of the frames. 

 It should have shown, if they were present, the effects of the cluster 

 on the temperature of the air below the frames. It might be expected 

 that the presence of the bees would have raised the temperature of 

 the air in this part of the hive. For comparison with the other tem- 

 peratures, thermometer o was hung in the shed in which the experi- 

 ments were conducted, and registered the temperature of the air 

 which enveloped the hive. Comparison of the readings of thermome- 

 ters/and o reveal some significant facts not altogether in accord with 

 the general belief of beekeepers. 



During the winter as a whole these thermometers registered almost 

 identically. Slight variations occurred, but only for a few hours at a 

 time, and may be attributed to minor influences of the cluster, to 

 peculiar atmospheric conditions, to drafts, and to the agitation of 

 the bees. It should also be noted that the air which came in the 

 entrance entered from outside the shed and the temperature of this 

 air may not have been exactly that recorded by the thermometer o. 



During the period of most protracted cold, from January 23 to Feb- 

 ruary 1, when the outside air ranged about 0° C. (32° F.), thermometer 

 / followed the outside temperature closely, and the course of the two 

 curves is practically the same. In some cases, as for instance on 

 January 26, thermometer f was slightly lower than the record of the 

 outside air, which may possibly be explained by lack of ventilation or 

 stagnation of the air of the hive. The lowest recorded outside tem- 

 perature was —10° C. (14° F.). Since it was impossible to read 

 these low temperatures on instrument/, and since the two curves are 

 parallel so far as records were possible, it may be assumed that ther- 

 mometer/would have registered almost the same as thermometer o. 



During the warmest days and nights the recorded temperatures 

 were the same. The maximum for the winter period came on March 



