146 THE. HUMBLE-BEE vil 
some species they are liable to great variation, and 
more confidence is to be put in differences in struc- 
ture, which are perfectly trustworthy when sufficiently 
great, although, unfortunately, they are not very 
numerous nor conspicuous. 
Some of the species differ from one another in 
the length, in proportion to width of the head, 
more particularly of the cheeks. In the males 
important differences are often shown in the length 
of the antennz, in the comparative lengths of the 
3rd, 4th, and 5th joints of these, and in the shape 
of the hind metatarsi. 
Many of the most closely related species differ 
in the quality of their coat; thus the hairs may be 
short or long, equal or unequal in length, dense or 
thin, fine or coarse, erect or decumbent. In some 
cases these differences are so slight as to be only 
noticeable when combined with one another. When 
the hairs are unequal in length the coat is said to 
be uneven or shaggy. 
One can always tell to what species or group of 
species a male belongs by examining through a 
lens the shape of the forceps and wands of the 
genital armature. An outline drawing of the arma- 
ture of each of the British species is given on 
Plate VI. In comparing specimens with these draw- 
ings it must be borne in mind that the slightest 
change in the position of the armature, or of its 
parts, alters its outline, bringing fresh parts into 
view and causing parts previously seen to disappear, 
and also that the markings seen on the upper sur- 
