7 July 2007

Ms. Christina Scheltema, CRM

Office of Pesticide Programs

U.S. Environmental Protection Program

Room S-4900, One Potomac Yard

2777 South Crystal Drive

Arlington, Virginia 22202-4501

Dear Ms. Scheltema,

I write to offer my review of the case study report on honey bees and
Carbaryl (MRID 47092001) conducted by Bayer Crop Sciences.  The study
examined the effects on honey bee colonies of Sevin XLR Plus applied at
the maximum label rate to poplar plantations.  The parameters measured
in the honey bee colonies were appropriate but I feel the study could
have been improved on the monitoring end with the following: 1) The dead
bee traps were not quantified by using marked dead bees to give a
recovery rate.  Without calibration these traps may be capturing only a
small portion of the dead bees in each colony. It is common practice to
place 20-50 marked dead bees in the hive at the beginning and end of the
study to calculate a recovery rate for each trap. 2) The 30 second
flight counts were too short and create too much variability between
counts for the counts to be meaningful.  Counts of 1-3 minutes are
needed to get a reliable estimate of forage activity.

The study in my opinion suffers from one major flaw; the plot size used,
1.4ha, represents only 0.2% of the potential flight area visited by the
colonies and as such one would not expect negative effects as bees would
visit other untreated areas and be able to avoid the sprayed areas.  The
estimate of 0.2% treated area is based on bees using a 3km diameter
circular area to forage in.  The untreated control area had a much
larger planting of poplar trees as indicated by the photos.  The treated
plot had a more diverse array of plants for the bees to forage on and
thus avoid the spray area. This is corroborated by the significant
increase in honey production in the treated plot compared to the
untreated plot (see Table 4.2).  The colonies in the treated plots were
not exposed to the treatment in a meaningful way.  I do not know the
average plot size that would be sprayed in a normal plantation but I
must assume it is greater than 1.4ha.  I would have more confidence in
the findings if the plot size were increased and the vegetation types
between the plots standardized (I realize it is hard to create identical
plots).  The small plot size relative to the overall forage area of an
average hive give me little confidence in the data presented.  This
study is not a rigorous test of honey bee potential exposure to Carbaryl
used in poplar plantations.  

Jeff Pettis, Research Leader

Bee Research Laboratory

Bldg. 476; BARC-East

Beltsville, MD 20705

301-504-7299

jeff.pettis@ars.usda.gov

United States Department of Agriculture

Research, Education, and Economics

Agricultural Research Service

Bee Research Laboratory

Building 476, Beltsville, MD 20705

An Equal Opportunity Employer

