With regard to previous press releases on confirmed reports of results of 
nuclide analyses at Fukushima Daiichi Nuclear Power Station, we have 
confirmed that there were some incorrect descriptions and therefore we 
hereby announce results of corrections. We will describe corrected 
information and how many corrections we have made as follows. 

<Corrected information (the number of corrected parts)>
·Errors on density limit by the announcement of reactor   regulation and 
 half-life (31)
·Errors on measured value (43)
·Errors on scaling factors (12)
·Errors on a unit in measured values (2)
·Errors on calculation formula (6)

 The number of corrected parts we had confirmed by August 30 was 94 in 
 total. (Previously announced on August 30, 2011)

Based on this matter, we received verbal directions from Nuclear and 
Industrial Safety Agency (NISA) of Ministry of Economy, Trade and Industry 
on the investigation to determine causes, the development of preventive 
measures and the review of nuclide analyses previously announced on August 
30, 2011. 

We seriously accepted NISA's directions with sincerity and reviewed all 
values of results of nuclide analyses as object of investigation, 
investigated to determine causes and develop preventive measures. 
The followings are the details of causes and preventive measures.

(Result of review of all values)
○Addition of corrected information after we announced on August 30
 -Incorrect description of density limit by the announcement of reactor 
  regulation (1 part was newly added)
  *corresponding to (1) in the corrigenda
 -Errors on measured value (3 parts were newly added)
  *corresponding to (2) in the corrigenda
 -Errors on scaling factors (12 parts were newly added)
  *corresponding to (3) in the corrigenda
○Result of nuclide analyses in other than confirmed reports (Pu and Sr in 
  the soil)
 -Errors on transcriptions of results of measurement and calculation 
  sheets of evaluation (11 parts were newly added)
  *corresponding to (2) in the corrigenda
○Corrected information on measurement conditions 
 -Incorrect description of time of sample collection (1 part was newly 
  added)
 -Describing ways of units (16 parts were newly added)
 -Incorrect description of directions of places of sample collection (2 
  parts were newly added)
  *corresponding to (6) in the corrigenda

 Due to the result of review of all values, we have corrected 140 parts in 
 total including ones we announced August 30.

(Causes)
 -We did not adequately confirm every time we should have done so or make 
  final confirmations before announcements. Hence, we described incorrect 
  values when we transcribed results of measurement in publicly announced 
  materials.
 -We have not confirmed errors in making calculation sheets and formats of 
  publicly announced materials such as making calculation sheets for 
  evaluation of nuclide analysis results, incorrect description of density 
  limit by the announcement of reactor regulation, half-life and 
  measurement conditions and continued using them.

(Preventive measures)
 -We have applied and will apply review sheets for confirmed reports to 
  results of nuclide analyses for other than confirmed reports regarding 
  samples collected on and after May 8. We will make clear items to be 
  reviewed including parts not directly relating to measured data and 
  confirm final reviews are conducted on a daily basis. In addition, 
  regarding confirmation by review sheets, we will set up columns for who 
  confirm, modify or make a final confirmation and make clear who is 
  responsible.
 -We will set up standard formats for calculation sheets and publicly 
  announced materials. Those who make materials are prohibited from doing 
  so by photocopying pervious materials and should make them based on the 
  formats in every time. We will take measures against prevent 
  modifications by electronically locking inputting data in parts where no 
  modification is required such as half-life and density limit by the 
  announcement of reactor regulation.
 -We will discuss with external analysis institutions their direct 
  description of measured results in publicly announced materials, as no 
  transcription is required. 

We will share these preventive measures and this matter with all people 
relating to nuclide analyses and handle them together to make them 
effective. 

<Attachment>
 -Series of previously announced object materials on nuclide analyses*
                                                           (PDF 285KB)
  *Excluding corrections we announced on August 30