263 
the smooth surface of the glass, and it at once runs up. 
If a glass is used, as I have described, at least ten speci- 
mens can be taken in it before they attempt to find their 
way out, when taking a new specimen. 
Great patience is required both with sweeping and 
examining the net. A hurried look gives nothing, a care- 
ful examination of 10 to 15 minutes gives one or two 
specimens. As the specimens are caught in glasses that 
cannot contain poison, they must always be brought home 
alive, which in itself can be an advantage if the living 
specimens should be examined. I preserve my specimens 
in small glass tubes (40 X 10 mm) in129(75%) spirit, which 
gives the advantage that they do not become stiff. These 
small tubes, which are closed with good corks, are then 
placed in a larges glass bottle with ground glass stopper, 
making it easy to renew the spirit. The wings, antennæ 
etc. necessary for drawing, are prepared in canada balsam. 
To make a collection of such minute insects, set on card, 
is quite hopeless; they would in most cases be completly 
drowned in the gum, pubescence distroyed andt he abdo- 
men shrivelled so that an idea of what the originally 
looked like is impossible. To sketch the specimens would 
also be out of the question, as it is only possible to obtain 
absolute accuracy when for example the wings are laid 
quite flat. Even a superficial microscopical examination is 
only made with difficulty when only mounted specimens 
are at hand. 
Some of the species are most probably bound to one 
special host, whilst others attack various hosts as previously 
mentioned with regard to 7richogramma, but unfortunatly 
very little is known on this point. Similarly how many 
parasites can live in a single host is neither clear. I am 
of the opinion however, that with the case of Trichogramma 
on the eggs of Sialis only one is bred from each egg 
as is most probably the case with Ophioneurus signatus 
on the eggs of Rhynchites betulæ, but on the other hand 
