This was first generated by OCR from the PDF, Then edited by jhs@ Not changing words, but just OCR tech corrections, namely: - OCR doesn't recognise clustered header blocks, to place in contex. - OCR failed to read oblique faint date stamp. - OCR inserted a few spurius blank lines in wide spaced original. - OCR doesnt know to strip inter page continuation indicators "./. -2-" - OCR doesnt know a multi page letter, received with only first page printed on headed paper with header & footer, when converted to computer text, is better with the footer delayed until final page. - OCR failed to insert dots on footer "www."