Mr. William Gardner
Secretary of State
State House, Room 204
Concord, NH 03301
We are hereby requesting that based on the following information and evidence, you use your constitutional authority to order a complete and accurate recount of all ballots in the New Hampshire Democratic Presidential Primary election.
Above and beyond any concerns for the integrity of the physical chain of custody of the New Hampshire 2008 primary election ballots prior to the recount, there are significant percentage variances in four voting districts in Hillsborough County.
In Wilton, stated by the New Hampshire Secretary of State's web site to be a hand-count voting district , a hand re-count revealed 100 fewer votes for Obama than originally reported on January 10th, a variance of 10.6% of the total votes cast.
The following three voting districts in Hillsborough County with large variances were originally tabulated on Accuvote optical scan voting equipment:
- in Nashua Ward 5, a hand re-count showed five Democratic candidates with a total of 108 total votes fewer than reported on January 10th, a variance of 4.9% of the total votes cast
- In New Ipswich, a hand re-count showed 50 more votes for Richardson than reported on January 10th, a variance of 7.5% of the total votes cast
- In Manchester Ward 5, a hand re-count showed five Democratic candidates with a total of 151 total votes fewer than reported on January 10th, a variance of 10.6% of the total votes cast.
An online news story at WMUR  attributes the Manchester Ward 5 variance to a manual error by the town clerk, who reportedly accidentally added votes from the ward's Vice Presidential contest to the Presidential results.
As of 10:00 PM EST on January 20, the New Hampshire Secretary of State reports 462 Democratic Vice Presidential primary votes were cast in Manchester Ward 5 . However, the method by which 462 votes for two Vice Presidential candidates could accidentally have been recorded as 151 extra votes for 5 Democratic candidates for President within the same ward through a manual error was not further explained.
The magnitude of the variances in the four voting locations raise questions about the integrity of the internal vote accounting procedures in use in Hillsborough County. A reconciliation of the records of votes cast to voters who signed in and received ballots should have detected these problems; why multiple locations with large over-voting discrepancies went undetected in Hillsborough County remains unexplained.
Given the narrow margins of the statewide outcome, manual verification of the remainder of the vote in New Hampshire seems warranted. Whenever there are indications of over-voting, an accounting of spoiled and unused blank ballots is also justified. We await your thoughtful consideration and determination.
Dennis J. Kucinich