Levelling the Rock.                                  

Those of us who enjoy walking around The Rock will have noticed various Ordnance Survey benchmarks, incised at specific points around our domain. They tell a story which started with the French Revolution and continued until the last traditionally-cut arrow-style benchmarks were carved in the UK in 1993.

The wars with France following on from their 1789 revolution, and the threat of invasion of Britain, stimulated detailed topographic mapping of the entire UK for military purposes. The task was funded and directed by the Board of Ordnance who established the Ordnance Survey of the Board of Ordnance in 1791 under Major Edward Williams, a Royal Artillery officer. The first topographical map produced by the new Survey (in 1801) was a one-inch-to-one-mile map of Essex, the county most likely to face invasion from France. By the 1840s the requirement had changed to 25 inches-to-the-mile as that had proved beneficial since its first use in Ireland in 1820s.  It was to be sixty years before the O.S. turned their attention to Gibraltar and that was thanks to a man who had been born on the Rock, on the 7^th January 1802.

            

                                    Detail from the first O. S. of the Rock; repeatedly updated and still in use today.

Edward Charles Frome, son of a military chaplain, was orphaned aged two, (by the yellow fever epidemic of 1804) received his early education at Bexley and Blackheath and entered the Royal Military Academy at Woolwich in 1817. He was commissioned a 2^nd Lieutenant in the Royal Engineers in 1825 and two years later posted to Canada where his work on the Rideau Canal, from Kingston to Ottawa, included a dam and three locks at Kingston Mills. This is now a National Historic Site of Canada and was designated a UNESCO World Heritage Site in 2007. Returning after completion of the canal, in 1833, he became an instructor of surveying and engineering at Chatham.

Among his many published works;  An Outline of the Method of Conducting a Trigonometrical Survey, for the Formation of Geographical and Topographical Maps and Plans, was to become the bible for Royal Engineers and others and remains relevant today.

In 1840 Captain Frome was appointed Surveyor General of South Australia and took with him a party of Sappers with orders to survey town sites, country properties and roads in advance of the arrival of settlers, so that the colony would be settled in a methodical and rational manner. As Colonial Engineer he supervised construction of many vital civil works. He was also responsible for surveying most of the known parts of the country and exploring vast tracts of unknown territory.  After returning to England he was appointed, in 1852, Surveyor General of Mauritius and subsequently, in 1859, promoted Colonel and Commanding Royal Engineer, first in Scotland, then Ireland and eventually in Gibraltar from 1862-66. 

Frome was an experienced and determined engineer. One of his first determinations was that a proper survey of Gibraltar was needed and he enlisted the help of a skilled and enthusiastic Royal Engineer, Lt. Charles Warren, who had arrived in Gibraltar from Chatham during the previous year.

                

                        R.E.s on the Rock 1863: Lt.Warren back row rhs, Col. Frome front row with banded cap.

Warren was a remarkable young man. Born in Bangor, North Wales, he was the son of Major General Sir Charles Warren, (1798-1866) who had marched on Paris with Wellington after Waterloo, served in China, Africa and eventually Gibraltar 1851-54, before commanding the 1st Brigade at Inkerman (Crimean War). Mentioned in Lord Raglan’s dispatches for being wounded ‘whilst leading his men with conspicuous bravery’, he was awarded a medal with clasps for Alma, Inkerman and Sevastopol. It was said that he had a natural turn for science and mathematics and his memory was so good that ‘he could hold in his mind all the figures of a long calculation and could correct and alter those figures at will.’ Many of these characteristics would be inherited by his son, Charles, born in 1840.

The younger Charles Warren was educated at Bridgnorth and Wem Grammar Schools, before attending Cheltenham College, from which he went to the Royal Military College, Sandhurst and then to the Royal Military Academy at Woolwich. He was commissioned as a 2^nd Lieutenant in the Royal Engineers in 1857. After the usual course of professional instruction, Warren was sent to Gibraltar in 1861, where he was to spend the best part of the next five years. His duties, as an Engineer subaltern included looking after his men, constructing or improving fortifications and building barracks. He was employed for some months in scarping the east face of the Rock to deny any possible foothold to an enemy. Never idle, Warren also designed a metal replacement for the wooden truck-levers used to adjust the traverse of cannon. However, his most important task was the trigonometrical survey of Gibraltar required by his boss, Colonel Frome.

Warren executed his survey on the grand scale of 100 inches to the mile (1:600) which compared favourably with the 25 inches to the mile used elsewhere in the military world.  Just for comparison, the scale we most commonly employ today is 1 inch to the mile (1:5000). His survey was completed with two 26 foot models (8’8” wide) of the Rock, one of which used to reside in the Rotunda at Woolwich and the other remains in the Gibraltar Museum. These highly detailed works show not only every road and lane but every individual building extant in 1865. This remarkable achievement, whilst designed for military purposes, gives us an amazing insight into the Gibraltar of those years and remains a valuable tool for historians, academics, geologists and geographers of today.

             

       Detail of Town area and Rosia from Warren’s model; pics courtesy of Ron Schantz.

 

How Warren conducted the survey is very interesting. The process was based on triangulation, a means of determining the location of a point by measuring only the angles to it from known points at either end of a fixed base line and so constructing a triangle from which additional triangles can be formed to extend the survey. By repeating networks of triangles, accurate maps could then be produced.  The fixed base would start with a 100-link survey chain used to establish the base line and theodolites to measure the angles.

Chains had been in use since the early 16C. but in 1593 after Queen Elizabeth 1^st redefined the English statute mile as 5,280 feet; the Old English mile based on the Roman Agrippa’s computation of an Imperial mile as 5000 feet (mille passus 1000 paces) and other variants all had to disappear. By 1620, the polymath Edmund Gunter had developed the most accurate method yet of surveying land by employing a chain of 66 feet long, (22 yards) using 100 links. It took fifty years to be fully accepted but Gunter’s chain became the standard. Ten chains made a furlong (furrow length) and eight furlongs made one statute mile. This standard remained in place, eventually crossing the Atlantic to the colonies and is still in use today.

                    

                                    Ramsden’s Engineer Chain.                                Surveying Staff graduations.

For the military surveyor, however, the need to survey greater distances led in 1784, to R.E. General William Roy, (famous for his survey of Scotland and a precursor of the Ordnance Survey) to commission one Jesse Ramsden to manufacture The Engineer’s Chain. This device was 100 feet long, simplifying calculations especially when used in multiple measurements. It was also constructed in a military manner; made of solid steel bar with double jointed hinges. Every 20^th link the hinge is set at 90 degrees to the normal, allowing it to be folded up into a neat, compact bundle of 5 sections each of 20 feet. Ramsden was a well known instrument maker who designed highly accurate dividing engines, telescopes, sextants and theodolites capable of measuring angles with an accuracy of half a second of arc.

These then were the instruments, which over the succeeding years were progressively improved upon and gave the Royal Engineers the tools to produce exceedingly good maps.

     

         Elliot Bros. surveying compass 1860s.       Cooke & Sons survey & leveling engineer telescope.

The chain and theodolite tell only part of the story. Firstly, the triangles are not really plane triangles. The curvature of the earth, even over the short 3 miles of the Rock, introduces an error of nearly 6 feet. Warren then had to calculate spherical triangles to produce an accurate result. Next he had to deal with the levels of the different areas or features of the land and delineate contours that reflected the actual terrain before him. This in turn would generate problems when it came to transposing the contours to the model.

To assist him in determining levels, Warren used a variety of kit. Initially, he used Mr. Gravatt’s level, also called the Dumpy level, so called from its shape, having a large aperture object-glass of short focal length, affording the advantage of a large instrument without the inconvenience of its length. The modern descendant of the Dumpy level may be seen any day at work in Gibraltar.  However, he also favoured the French Water Level, much used by continental military engineers, and had one constructed by an ironmonger in Chatham.

                 


                                               Dumpy level.                                                   French water level.

    

The principal advantage of the French Water Level is in not requiring any adjustment and ‘does not cost the 2oth part of a spirit level’.  Warren frequently used it to interpolate 25 foot contours in between the more correct contours traced at greater distance apart by the spirit level. Although not minutely accurate it was not susceptible to errors introduced by a badly adjusted spirit level or theodolite. In 1863 Warren subsequently modified it to improve its performance. Warren also employed the reflecting level invented by Colonel Burel and Elliott’s pocket reflecting level, both of which require some pages of explanation to do them justice.

 

                                

                                                           Colonel Burel’s reflecting level.

Warren’s survey was plotted in detail on 42” x 26” antiquarian paper plates at fifty feet to one inch (1:600). The model was constructed on the same scale and built on eight frames, 7’ by 4’ 4”, each frame occupied by four plates. These frames were formed of well seasoned 3” deal, dovetailed, screwed and braced so as to be perfectly rigid. Great pains were taken to ensure they were absolutely square and consequently the eight fitted together perfectly true. They were partially covered with a skin of ½” yellow pine, the surface of which is sea level. Onto this was transferred the coastline and outline of sea-level fortifications. The summit of the Rock at 1400’ above mean sea level resulted in the extreme height of the model being 28”. The inclination of the surface in some areas being virtually perpendicular, the ordinary system of modelling could not be carried out ‘but a more rigidly accurate method was adopted by means of the contour lines’.  The exact heights of places on the edges of the plates being known, the edges of the frames were continued in wood to their proper height, great care being taken forming the sections. Then planks of yellow pine were cut and planed to 0.4” each piece representing a height of 20’ and so five of them (2”) made 100’. Traces of the contours were then taken, and the lines transferred to the pieces of yellow pine, in duplicate; then taking a keyhole saw, the wood was cut through on one set at the contours 60, 100, 140, 180, etc and on the other set at 40, 80, 120, etc. The pieces were applied to the frame edges and glued up. For filling in between the contour lines (the problem mentioned earlier) tin tacks were nailed in differing directions on the pine surface before ordinary plaster of paris, mixed somewhat stiff, finished the surface.  It now only remained to give the geological features to the surface of the rock, to make the cuttings and add the excrescences, whilst cutting the general features on the steep portions. For this, photographs and sketches were used. 

The fine detail of forming the successive layers of contours was achieved by Sergeant Turnbull and Sappers Williams and McLellan, RE who are rarely credited for turning Warren's exacting survey into a credible model. Neither is Captain B A Branfill, who painted the model in 1868 and provided us with the exceptional finished model we enjoy today.

Warren’s subsequent career as surveyor of the Temple Mount, his part in the Transkei and Boer Wars, the annexation of Griqualand West and his role as the Metropolitan Police Commissioner, all deserve articles of their own but hardly relate to our Rock.

 

                                             

                                                                            Charles Warren C. 1886

First published at Insight Magazine 2001 - revisited March 2025.    Paul Hodkinson.