Roch Mill is located in a small picturesque valley close to the sea in west Pembrokeshire National Park. Its recorded origins date back to the 13th century when it belonged to Roch Castle, a Norman watchtower on the extremities of "English" rule. The castle was built by Adam de Roche (of the rock) and it is generally believed that Roche became shortened or anglicised to the present day Roch (pronounced Roach).
Situated inside the Pembrokeshire National Park and just two miles from spectacular beaches like Newgale, Druidstone, Nolton Haven and Solva harbour, the six acre Roch mill estate nestles in the secluded Brandybrook valley.
This pretty little river boasts a significance far greater than its size would suggest as it forms the western end of the boundary between the 'Welshry' and 'Englishry'. It is the historical Landsker Line and marks the limit of Norman influence nine centuries ago. Even today place names to the north of the line are nearly all Welsh whereas those to the south are predominantly English. Similarly the Welsh language is much more frequently heard towards St Davids and the Preseli Hills, north of the Landsker Line.
Though seldom following any distinct natural feature the Lansker Line is marked here and there by a series of 'frontier' castles or Watch Towers. Roch Castle, which overlooks the Brandybrook valley, marks the western boundary. It was constructed in about 1295 though the current owner (Keith Griffiths) thinks it might have been as early as 1195. First mention of Roch Mill occurs in the early 14th Century when it appears as part of the castle's estate. Thereafter Roch Mill is mentioned in castle records every time the castle changed hands or some other event occurred that required a legal record of castle assets.
Our historical research has uncovered quite a lot about the Mill's history but this is not the place for a detailed discussion.
However, there are a few snippets from the records that deserve some mention. Belonging to the castle gave Roch Mill and its tenant considerable status and, therefore, income potential. On 16th February 1733 Richard Griffiths took out a 21-year lease on Roch Mill, including 8 acres of land. The rent at that time was "Six fat hens and one green goose as duties or customs and the carriage of three barrels of culm from Roch parish to the town and county of Haverfordwest and also the yearly rent of £10.15s.0d".
[Note: Culm is a mixture of coal dust and clay. A combination that enabled the coal dust to be used effectively as a cheap fuel]
The Griffiths family became important participants in the history of Roch Mill. In 1814 David Griffiths, great grandson of Richard borrowed £400 to buy a 500 year leasehold (for the mill and 51 acres) and, remarkably, paid it back with interest (£650) only 5 years later. Clearly, the mill was very profitable. As the primary mill of the Roch parish it is likely that the land-owner required all of his farmer tenants to use Roch mill exclusively.
On a more personal note, by to my reckoning David Griffiths' lease still has 302 years to run and I wonder why this wasn't picked up by our solicitor when we bought Roch Mill!
In payment for his services it was common practice for a miller to demand a percentage of the grain presented. Typically this would be one quart per bushel or about 12.5%. Being in full control of the milling process it was not difficult for a miller to "adjust" this figure according to his likes and dislikes. It is hardly surprising that in some parts millers became synonymous with cheats and swindlers.
The Griffiths family owned Roch mill from 1733 until the late 19th century, about six generations, and were responsible for replacing the main machinery in 1868. Aged 71 in the 1891 Census Daniel Griffiths appears to be the last of the line.
At the head of this page is a banner copied from a splendidly illuminated document produced in the reign of George Second. Unfortunately, the document itself is not very illuminating! It is just a pedantic court record relating to Roch Castle and its possessions. Although the reference to 'two corn grist mills' is about as close as the 1735 deed comes to recognising Roch Mill it is a pretty safe bet that one of them was indeed our mill.
It is uncertain, however, just how much of the present building and machinery was in use in the early 18th century. From a date stamp on the main axle it is probable that most of the existing machinery was installed in about 1868. This coincides nicely with the widespread conversion from wooden to iron wheels and axles during the industrial revolution. The frame, or hursting, is believed to be much older.
The 'Hursting' or wooden frame that supports the mill wheels and machinery is clearly much older. Some iron nails have been dated as 17th Century or earlier. No doubt some of the building foundations are much older as there are references to a mill at Roch going back over 700 years. Interestingly, some of the supporting timbers in the mill are riddled with giant-sized woodworm holes, suggesting that they might been caused by tropical wood-boring beetles and the planks have been 'recovered' from a shipwreck on local beach.