6 BULLETIN 1222, U. S. DEPARTMENT OF AGRICULTURE. 
surrounded by food. All seemed large as measured by the eyepiec 
micrometer. One unhatched egg was seen. According to previous 
observations of the writer, these eggs should have hatched betweei 
11.15 a. m. and 1.15 p. m. on this date. On July 26, 5 of the 11 
larvae present were weighed, the individual weights in milligram; 
being as followed: 5.05, 6.1, 20.9, 23.4, 28.9. Considerable differ- 
ences in the size of the larvae were evident on examination of tht 
comb. Three of the larvae covered an area of less than one-half o1 
the diameter of the bottoms of their respective cells, but most ol 
them nearly or quite filled the bottoms of the cells. All of the larvae 
were supplied with an excess of larval food. 
VARIATIONS IN RATE OF GROWTH. 
An outstanding feature of these observations is the great individ- 
ual variation in the daily rate of growth as manifested in the dif- 
ferences between the weights of the 5 to 10 larvae of the same age 
in different lots, as well as in the differences between individuals of 
the same age in the same lot. These considerable differences natur- 
ally led to the suspicion that they might have been due to the method 
used in selecting newly hatched larvae. The experiment described 
under Lot 7 was intended to test this supposition. As already 
stated, larvae 3 days old, hatched from eggs laid within a determined 
3-hour period, showed wide divergences in individual weight. 
There were on hand two series of larvae, each from a single batch of 
eggs laid within a known period of 2 hours, preserved in alcohol. 
The greater part of these were fixed in their cells, the earlier stages 
being embedded in the coagulated larval food and therefore unsuit- 
able for weighing. Five larvae 3 days old of one lot were free from 
foreign material and suitable for weighing for comparison with one 
another. These showed on superficial examination considerable dif- 
ferences in size. These larvae were weighed separately, the weights 
in milligrams being as follows: 5.7, 7.35, 7.5, 9.8, 10.4. Although 
the differences in weight between the different individuals of this 
lot are not as great as between those of most of the other lots, still 
they are considerable, the difference between the smallest and the 
largest amounting to nearly 100 per cent. This result, taken in con- 
junction with the great differences between larvae of the same age 
in different lots, makes it reasonable to conclude that considerable 
differences exist in the rate of growth of bee larvae. 
In fly larvae Herms (2) determined that "there is an optimum 
when enough nourishment has been taken to pass through the 
metamorphosis to the best advantage." This optimum is evidently 
represented in the honeybee by a weight of about 158.31 milligrams 
when the larvae are sealed by the worker bees. As far as the pre- 
ceding observations on the honeybee are concerned, the optimum 
weight virtually coincides with the average weight at maturity. 
LENGTH OF THE LARVAL PERIOD. 
As the data show, sealing may begin as early as the end of the 
fourth day after hatching (Lots 2 and 3) and as late as the middle of 
the fifth day (Lots 1 and 5), the average for the five lots being about 
4r| days (Table 1). It is noteworthy that a difference of half a day 
may exist between two lots from the same hive, not only during 
