O. H. Latter 
203 
Type I Pure 50^ 
I + 11 111 62 
I + III iJ 
Type II II + 1 22^ 
Pure 94 [ 120 
II + III 4] 
Type III III + II n 
Pure 3 1 4 
III + IV oj 
Type IV IV + III 0^ 
Pure 1 [ 
IV + v oj 
Type V V + IV 0 
Pure 1 
Total: 188 
It will be seen that transitional cases undoubtedly occur. The bulk of the 
queens, however, fall into Types I and II, or queens are very little variable. 
To test : (i) whether this variability was still further lessened by taking only 
the queens from a single nest, and (ii) the relative variability of queens, drones 
and workers, I now examined all the queens, workers and drones of a single nest 
of V. vulgaris. 
In this case all the 127 queens were of Type II*. 
The classes of the workers are given in Table II (p. 205) and may be sum- 
marised as follows : 
Type I Pure 5\ 
V. s or (I + II) ? oj 
Type II II + 1 6 
Pure 156 
10 
162 
Total: 172 
It will be seen that they are somewhat more variable than the queens of the 
same nest, but not so variable as queens from different nests. 
I now turn to the drones of this same nest. I had 150 at my disposal. 
The drones exhibit a very wide range of facial markings. In the material 
examined comparatively few fall into the scheme of classification adopted for 
the queens and workers, and it thus becomes necessary to resort to six types 
of face which appear to be peculiar to the male sex. These are numbered VI, 
* There were 129 queens in this nest, but No. 34 was missing and No. 98 had its head damaged too 
badly for classing. 
