184 
The larva spends all its life below the surface of the water 
but is not provided with anj respiratory organs which might 
indicate that the animal is adapted to the aquatic life. 
Perris has shown that the larva does not rise to the sur¬ 
face for breathing, and so he thinks that it breathes the oxygen 
dissolved in the water. 
v. S i e b o 1 d is the lirst author who puts the respiration process 
in connection with the air in the intercellular spaces of the roots 
and rhizomae on which the larvae are living, and in this view 
nearly all the later Authors agree with him. 
In order to make clear in which way this must be thougt to 
take place I shall raention the following from Bovin g. 
The larva is provided with a so called abdominal organ or 
appendage which is homologous to the Hister spiracle. 
The abdominal organ consists of two distinet portions: the 
basal part and the hook-shaped part. 
The hooks are perforated with two pairs of canals; these 
canals communicate with the widened part (the atrium) of the 
abdominal organ x ) and from here there are passages into the 
tracheal system of the larva. 
When the larva therefore is sitting with the hooks stuck 
into the body of plants there is a direct communication between 
the intercellular spaces of the plant and the tracheal system of 
the larva. 
The animal seems therefore to be able to inspire the inter¬ 
cellular air through the hooks mueh in the same way as an 
ordinary larva inspires atmospheric air through its spiracles. 
The above mentioned explanation is not Bøvings but my 
*) The walls of the atrium are covered with emergences; at the spot 
where the lateral pair of the canals (it is only this pair that commu- 
nicates direetly with the atrium) join the atrium, the emergences 
are placed as a dense filter but form according to Schmidt-Schw r edt 
and Bøving no obstructing wall. 
