advances in the fields of machine vision
techniques, digital-image analysis, rapid
counting and sorting techniques and
quantitative immunological and molecu-
lar techniques could open up possibilities
for the development of new, less labori-
ous methods for quality and quantity con-
trol. In many cases, the results of quality
control tests become available when the
products have already been sold. This is
especially the case with products that are
sold as pupae (e.g. Encarsia formosa,
Eretmocerus eremicus, Aphidius colemani,
Aphidius ervi). Instead of performing
emergence tests, it might be possible with
some products to assess the viability of
the pupae at an earlier stage using bio-
chemically based techniques.- Developing simple testing methods and
standards for end-users. Most quality
control guidelines are too tedious to be
performed by growers. Therefore, simple
indicative tests need to be developed to
allow growers to quickly assess the qual-
ity of the products upon receipt.
However, it currently remains unclear
whether the results of such approximate
tests have sufficient power and accuracy
to allow for formal complaints and prod-
uct replacement or, worse, for liability
claims. The IOBC working group
focused on developing guidelines and
quality standards at the producers’ level.
It is still unclear whether these standards
are also valid upon arrival of the product
at the end-users’ facilities, especially for
parameters such as emergence rate,
fecundity, longevity, parasitism/preda-
tion rate and flight propensity. If the pre-
sent methods cannot be used, different
quality standards need to be developed
for products after shipping. - Developing testing methods relating lab-
oratory tests to field performance. It is
not clear to what degree the currently
used laboratory quality control tests give
sufficiently reliable information about
performance of the natural enemies in
the greenhouse or field. This subject is
discussed in Chapters 16 and 17 by
Steinberg and Cain, and Luck and
Forster, respectively.- Quality standards: minimum or average?
At this moment, there is no agreement
within the industry as to whether quanti-
tative product specifications are mini-
mum quantities or average quantities.
This issue has led to numerous debates. It
is well known that, for many products,
there is a rather important variability in
the number of natural enemies per bottle
due to the packaging techniques used. It
is my opinion that 95% of the bottles of
one batch should contain, as a minimum,
the number of natural enemies that is
stated on the bottle. For the remaining 5%
of the bottles, underpacking of more than
10% should not be acceptable. But it is
important to realize that putting an exact
number of natural enemies in a bottle is in
practice rather difficult. Automation of
the packing process based on volumetric,
weighing or ‘counting-and-diluting’ tech-
niques often leads to even higher variabil-
ity than manually pootering individual
insects into a container. More research
and development needs to be done on
mechanizing this very important stage of
the production process in order to get a
better grip on the packaged quantities
and therefore product quality. Packing
techniques with a small standard devia-
tion also require a smaller number of
samples to obtain a reliable estimate of
the average number of insects per bottle. - Harmonizing the developed guidelines.
The format of the tests, the timing and
frequency of the different tests and the
specified test conditions could use some
further harmonization.
- Quality standards: minimum or average?
From Quality Control to Quality
Assurance
The main focus of the activities of commer-
cial producers and scientists in the IOBC
working group has been on product quality
control. Quality control is a retrospective
activity that focuses on the quality of a prod-
uct at the end of the production chain.
Quality control does not contribute to prod-
uct quality unless action is taken when prod-
ucts are detected that do not meet predefined218 K.J.F. Bolckmans