53 
above it, in front of it, and laterally ; while the under ones make 
it possible for it to see at the same time directly downwards as 
it swims on the surface of the water. 
3rd. The structiire of the antennse is remarkable, and differs greatly 
from that of the Dytucidce, though it is very like what we find in 
Parnus. Each is inserted in a cavity at the side of the head ; 
the first joint is very small, the second is large and dilated, and 
the third, also large, is inserted at the side of the second, while the 
remaining joints are so compressed and soldered together that it is 
not decided whether the antennae consist altogether of ten or 
eleven joints. 
4th. While in the Bytiscidee the mesostemum is small and feeble, and 
the metastemum is largely developed, in the Gyrinid<£ the meso- 
stemum is large, while the metastemum is correspondingly reduced 
and small. 
5th. The structure of the legs is most remarkable in the Gyrinidee, 
and affords in several respects one of the most interesting examples 
of the modification of organs to serve special functions that could 
well be instanced ; while the four posterior legs are formed into 
powerful swimming organs, the anterior are quite different — they 
are elongate, and are so placed that they can be packed under the 
body so as to offer not the least impediment to the most rapid 
motion, while by one or the other being thrust out the course of 
the insect is instantly changed, or when both are thrust out 
retarded, and thus the Gyrini are enabled to perform those rapid 
and eccentric motions which have attracted the attention of all 
who have eyes and can use them. This rudder-like function of 
the front legs is also perfected by the peculiar position in which 
they are placed, a position so strange that what should be the 
under surfaces of the anterior tarsi look towards one another, 
instead of downwards : dependent on this is also a peculiar modifi- 
cation of the tarsi, which are compressed laterally, so that, not- 
withstanding the peculiar position of the legs, the broad aspects 
of the tarsi are presented upwards and downwards as in other 
beetles ; still stranger is the fact that what is in reality the side of 
the tarsus is thickly furnished in the male with pecuHar hairs such 
as are placed in other beetles on the real under surface of the tarsi. 
Had these hairs been placed in a position anatomically the same as 
they are in other beetles, they could have been of no use for the 
purpose for which they are intended ; thus they are in a position 
