1. Enlarged brood cells affect the size of the adult worker bee, and significantly larger worker bees are obtained through the use of enlarged cell foundation.

2. The average percentage increases of the linear measurements of the adult worker bees are almost proportional to the percentage increases of the diameters of the brood cells.

3. The increase in the size of the bee does not· quite keep pace with increases in cell size.

4. Size of brood cell apparently does not affect the variability of the adult worker bee, except possibly its dry weight.

5. The number of bees used in a sample in this experiment is not great enough to give wholly consistent results, but these results in general are significant and indicative.

6. Dry weight is difficult to measure and further experimental control is needed.

7. Among body measurements, exclusive of proboscis, length of right forewing gives the best estimate of length of proboscis.

8. Of the two major parts of the proboscis, the mentum (which is the more easily measured) is more highly correlated with proboscis length than is the glossa.

9. A combination of length of right forewing and mentum length affords an excellent estimate of proboscis length.