The Garw Valley is divided into the upper and lower sections by a geological feature known as the Moelgilau Fault Line and Felin Arw corn mill stood on this line. The mill was recorded seemingly well into antiquity, and appears to have been the only one in the upper section. The site lay behind what is now the Richard Price Centre (RPC), adjacent to the lane leading to Cwm Square. The only surviving remains are the cottage, formerly two that were converted; and some stonework from an out-building which formed part of a stable, (1997 visit). Remains of the mill itself do not show above ground, and it was further up-stream.<more!>
The site of the Oat Kiln is still subject to research. This necessary building, which was rush-thatched; appears to have been somewhere near to Cwm Square, and one old resident mentioned it having been by the old Ogmore & Garw Relay Station. This kiln having to be a two level building, drying floor above with the heating room below; would have been placed where split-level access was easy. The names Ffynnon-yr-Odyn and Cae’r Odyn Villa allude to the oat kiln’s existence.
The earliest surviving Land Tax Assessment, that of 1767; records that Melin Arw was owned by Lady Charlotte, the tenant was Samuel Price, and that tax was assessed at 11/7d. (57½p.).
According to an archaeological survey of Llangeinor Parish the original source of water to the mill was drawn from the river Garw. The water channel, or leat, which appears to have originated just a little upstream of Ffynnon Dwym; was later cut through during the construction of the Great Western Railway Garw Branch which opened in 1876. The only water channel shown on OS maps is at right angles to the flow from the spring at Ffynnon Dwym. There is nothing on 1877 map indicating the site of a weir or the course of the leat either side of the railway to the above channel. The survey continues by saying that an alternative source of power was then taken from the water falling from a series of springs on Llangeinor Mountain down through Cwm Felin Arw.
Unfortunately the survey does not explore the human history of the mill, and thus the question arises in my mind as to the above change in water source. (See below). It seems from OS Maps that the original course of the mountain fed brook meandered between the Mill and the RPC before being conducted under the road and discharged into the river. The present day open water course runs on the opposite side of the lane, before doing an almost right-angled turn to join its original course to the river. This suggests that the mill wheel was on the side facing up valley, and the open water is the tail-race, that is to say the water which was left after turning the wheel.
At some time prior to 1870 the building was enlarged, a woollen factory and a bigger water-wheel installed. The story told was that the building had an extra floor built on to take the machinery, which cost £220; and took three days to arrive from Gloucester by horse & waggon. With the mill already being a multi-storied structure, it opens to question whether or not the changes were to the machinery, rather than to the dimensions of the building. The aforementioned survey does not include the factory in its paper.
The OS map shows a Bench Mark of 285⅓ feet near where the mill lane joins the Bettws road. The same map shows a Spot-height of 310 feet on the railway bridge at the first crossing of the river above the mill. Any weir/leat connected to the river was probably lower, say at 305 feet. This would have provided a fall of about 20 feet above the mill. Provided the height of the water was at least 10 feet at the wheel then it would strike above the axle. There is a great disadvantage of using undershot, where the water strikes below the axle, in that power is lost by the wheel having to push the used water away from itself.
The alternative, as above, were the springs on Llangeinor Mountain which lie between 600 and 900 feet. By having a shute taking water from near Cwm Square thus raising the level at which the paddles were struck, and with the increased the head of water; it would have given greater power for the machinery. To give a modern approach to the subject, imagine a toilet cistern at seat level, rather than at chest or head height; one can visualise the difference in pressure available to flush.
As a point of contrast, the mill at Pandy which stood no higher in relationship to river level than Felin Arw; took its water from behind a weir at least ½ mile upstream. Should the same length of leat have been required to supply Felin Arw then a weir would have been about the site of the Black Bridge over river and railway just below Tylagwyn. Any weir could not have been much higher than 324 feet, the railway line having a spot height of 329 feet just before the Black Bridge.
The problem exists of estimating the flow of water coming down the river fed leat when compared with that sourced from Llangeinor Mountain. The volume of river water, depending on the dimension of a weir/leat; may have been un-reliable throughout the year especially during dry months. In bad weather flooding would have been a major problem and no gate on the leat could have stemmed the flow. Maps suggest that water from Ffynnon Dwym instead of being allowed to flow straight down hill into the river, was diverted along to just above the mill. This could have been used to water meadows during dry months. As in the case of Pandy, the surrounding land height of that location, and the minor flow of water from the same, when compared to that at Felin Arw, clearly shows the need for a leat from the river.
A personal examination of the Tithe Map of Llangeinor parish of about 1840, does indeed show the same line as on the OS map of 1877, and this earlier line does appear to connect to the river slightly above Ffynnon Dwym, however no indication as to it being a water course is given.
If we allow 3 years for the GWR to construct their line from Brynmenyn until its opening in 1876, then the destruction of any leat from the river could not have taken place much before 1873. The last man known to have milled at Felin Arw died on Christmas Day 1870. However there is no indication of a date for when milling ceased prior to his death. None of his five sons or four sons-in-law have been found recorded as a miller.
There were weavers listed in 1851 census in (Cwm) Felin Arw, however they had moved away by 1861. Only one son of the last miller was recorded as a weaver in a woollen factory and that in the 1871 census; however this weaver had died by1875.
There has been no clear indication as to when the mill/factory fell into decline. The census of 1881 does not list either milling or weaving, and the earliest OS map of 1877 only records the site as being a corn mill, with no reference to there having been a factory.
If, taking into consideration that the deaths of the last miller was in 1870, and the last known weaver in 1875; would Lord Dunraven have taken the trouble to source water from elsewhere for a business that was not only in decline, but possibly having closed altogether? However the question of the continued use of the buildings for industrial purposes after the 1873-1876 construction of the railway still needs an answer. When it came to civil engineering, the GWR could have easily have built a conduit under the railway to maintain supply if the mill etc. were still in good use, but it appears not to have been. (I may have now answered my own question!)
In conclusion I must honestly confess that my earlier opinion as to the river fed mill leat being questionable now seems to be answered.
C T Davies (2016)