185 
own interpretation; according to Bøving the canals of the hooks 
are closed outwardly with a very thin membrane permeable for 
air. The membrane, however, should be so thin that it has not 
been possible to see it. 
If the canals are closed with a membrane the larva can not 
ventilate its tracheal system by active inspiration; the ventilation 
Can take place only by diffusion. As it is very difficult to under- 
stand the respiration of the larva without active inspiration — as 
will be shown later on — it seems to me legitimate to deny the 
existence of the membrane. 
Besides this explanation there is another, which (under different 
forms) has been maintained. by several Authors (v. Siebold, Mac 
Gilliwray and Deibel”)). Briefly expressed the theory amounts 
to this, that the larva catches the air, which is thought to stream 
out from the intercellular spaces when it sticks the hooks through 
the wall of the roots, by means of the .spiracular split which is 
lying at the base of the hook and opens into the atrium. (Accor- 
ding to Bøving this spiracular split only serves for expiration). 
This interpretation runs against facts; the intercellular air 
will not stream out when the wall of the intercellular space is 
perforated, as direct observation shows. 
If the animal makes use of the air in the intercellular spaces 
this will only be possible by supposing an active inspiration 
through the canals of the hooks. : 
We must be aware, however, that since no exact proof has 
been given of the necessity of the larva breathing the inter- 
Cellular air?), there might be the possiblity of the animal being 
provided with cutaneous respiration, as Perris has assumed %). 
7) Beitråge zur Kenntnis von Donacia und Macroplea. Zool. Jabrb. 
Abt. f. Anat. 1911. 
”) Babak writes (in Wintersteins Handbuch Bd. 12 p. 471): "Es bleibt 
also — scheint mir — von der physiologischen Seite der ganze 
Sachverhalt bisher vållig problematisch”, 
5) Even nowadays this interpretation has been. maintained (Sanderson). 
