On 5 September 1538, following the split with Rome, Thomas Cromwell, Henry VIII's Vicar General, ordered that each parish priest must keep a book, and that the Parson, in the presence of the wardens, must enter all the baptisms, marriages and burials of the previous week. The book was to be kept in a "sure coffer" with two locks (one key for the vicar, the other for the wardens). A fine of 3s 4d was to be levied for failure to comply. Many parishes ignored this order, believing it to be the forerunner of some new tax.

The order was repeated in 1547 with the stipulation that the fine was to go to the relief of the poor. From 1598, records were to be kept in 'great decent books of parchment' and copies or 'Bishop's Transcripts' of new entries were to be sent each month to the diocesan centre. Previous records (especially from the first year of Her Majesty's reign - 1558), often on scraps of paper, had to be copied into the new books, but many had deteriorated and were unreadable. The costs of the new books were to be met by charging for entries; this was opposed by many parishes and the act was not enforced until 1603. Finance was to be borne by the Parish, and the books were to be kept in a chest with three locks. The week's entries were to be read out each Sunday after evensong.

During the English Civil War (1643-1647) and in the following Commonwealth period, records were poorly kept and many are now missing after being destroyed or hidden by the clergy. During 1653-1660 the registering of births, marriages and deaths was taken over by civil officers (confusingly called Parish Registers), but the registers were returned to the churches following the Restoration in 1660.

I am sure most researchers will have come across entries from Parish Registers being transcribed in an ambiguous format, such as 6th February 1702/1703.
If you don't know the reason for this, read on.

When I came across a baptism on 29th February in a non-Leap year, I just had to check this out. The House of Commons Journal for Wednesday 29th February 1659 would seem to bear out this possibility, although Samuel Pepys diary has an entry on 29 February 1660 (just to add to the confusion) - presumably this would be transcribed as 1659/60.

This can all be explained by differences between the Gregorian calendar and the Julian calendar.

The Gregorian calendar was a reform in 1582 to the Julian calendar. The reform was adopted initially by the Catholic countries of Europe. Protestants and Eastern Orthodox countries continued to use the traditional Julian calendar and adopted the Gregorian reform after a time, for the sake of convenience in international trade.

Britain and the British Empire (including the eastern part of what is now the United States) adopted the Gregorian calendar in 1752. The last European country to adopt the reform was Greece, as late as 1923.

Under the Gregorian calendar February has 28 days unless the year is divisible by 4, when it has 29. However, if the year is divisible by 100 it gets more complicated - February only has 28 days unless the year is divisible by 400. So 2000 was a Leap Year, but not 1700, 1800 and 1900.

However, in the Julian calendar, an extra day was added to February every 4 years. As a result, the calendar year gained about three days every four centuries, a discrepancy corrected by the introduction of the Gregorian calendar.

In England, prior to 1752, the year began on 25th March (Lady Day). So in 1700, the month that other countries were now calling February 1700 was actually February 1699 in England. So February 1699 had 29 days; February 1700 had 28. Similar logic applies to the example of 1659 quoted above.

Prior to 1752, Parish Registers reflect 25th March as being the start of the year and is the reason why dates prior to 1752, which fell between 1st January and 24th March, being transcribed in an ambiguous format, such as 6th February 1702/1703.

By the time Britain adopted the Gregorian calendar in 1752, the Julian and Gregorian calendars were 11 days out of step.

The Calendar Act of 1751 changed the first day of 1752 to 1st January, technically known as Supputation of the Year. As a consequence, 1751 was a short year, as it only ran from 25th March to 31st December.

To bring the two calendars into line, the Act also deemed that Wednesday 2nd September would be followed not by Thursday 3rd September, but by Thursday 14th September - so September 1752 had just 19 days.

Lady Day was one of the days when rents and tax traditionally became due. It seems for a while at least, the British tax year continued to reflect the Julian Calendar. From 1753, the tax year began on 5 April, which was Lady Day adjusted for the 11 lost days of September 1752. A 12th skipped Julian Leap day in 1800 changed its start to 6 April. This date was not changed when a 13th Julian Leap day was skipped in 1900, so the tax year in the United Kingdom still begins on 6 April.