254 On Aquatic Carnivorous Coleoptera or Dytiscide. 
so articulated that it undergoes a process of semirotation when moved through 
the water, as the result of which when acting as a propeller it presents a broad 
face to the water, but when it is passive or is moved through the water for the 
purpose of attaining a position from which to commence another stroke it presents 
an edge ; in other words, what rowers call feathering the oar is performed by the 
tarsus of the Dytiscidee in a most perfect manner. Both the true upper, and the 
true lower, surfaces of the tarsus are mere edges, and the former bears elongate 
swimming hairs, which are depressed when the limb is at rest or moving 
forwards, but are spread out when the foot is acting as a propeller ; the latter or 
lower edge being closely set with spines. In Pelobius, however, the sole of the 
tarsus still exists as it does in the Carabide ; that is to say the lower face of the 
foot bears two series of spines and between these series a considerable space 
intervenes. It is an interesting fact that in many of the higher Dytiscide there 
are remnants of this structure still to be seen in the highly changed tarsus; the 
sole it would seem, disappears, at any rate in some cases, not by any diminution or 
degeneration but by a process of growth of the tarsus and concomitant modification 
of form by means of which the sole becomes part of the inner face ; in other words 
the inner face of the tarsus in the higher Dytiscidee is homologous with both the 
inner and lower faces of the tarsus of Carabidee. In the Carabide the tarsus has 
two sides curving into one another so that an arch forms the upper portion of the 
tarsus, and a sole, this latter being bounded on each side by a series of spines. In 
Pelobius the two sides of the Carabid tarsus and the sole still exist, as in the 
Carabidee, but form a triangle, of which the outer face is the longest side, and the 
true inner face of the tarsus is quite narrow, not so broad in fact as the sole, 
which is still marked off by a series of spines extending the whole length of the 
foot; in Dytiscus Sturmi (Agabus No. 737) where the tarsus is thoroughly 
Dytiscideous in form, there may be seen on the inner face of the basal joint a 
series of three or four spine bearing punctures, placed at a distance from the edge 
and evidently representing a remnant of the series of spines which serves as the 
inner boundary of the sole in Pelobius and the Carabidee; in Scutopterus there is 
a similar series of punctures on the first and second joints; and in the interesting 
Australian genus Hyderodes, this series of spine-bearing punctures actually still 
exists along the whole length of the four basal joints of the tarsus, on its inner 
face. Besides these spines which exist occasionally, there are on the inner side 
of the tarsus, other spines more constantly present; these consist of three or 
four spines placed on the hind margin of each joint, near the lower edge, and of 
some others similarly placed close to the upper margin; in Cybister the lower 
of these series of spines is reduced to a pair placed on each of the second, third and 
fourth joints; in Laccophilus this series is entirely absent, and the upper series 
consists of a pair of suberect spines, on the first, second and third joints, one of which 
being much longer than the other, projects beyond the lower margin so that its 
