Much focus in the historical railway literature has been placed on the role that the state played in imposing safety measures on the Victorian railway companies. Less light has been shed on what the railway companies themselves did to improve the safety of the industry for both the passengers and their employees.
Towards the end of the century many medical professionals became aware that some individuals’ colour blindness was potentially a danger to the lives of the travelling public if they worked in safety-critical parts of the railway industry. By the 1850s most railway companies used coloured signals to stop and start trains. Thus, an individual who was colour-blind may not have been able to distinguish the signals’ colouration and potentially caused an accident if he misread one at danger. Indeed, this was a greater problem at night as most railways used identically sized circular light signals. In 1857 The Derby Mercury reported a lecture by Professor George Wilson. In it he stated that:-
“people who are colour-blind are generally insensible to red and green colours, or rather confound these colours when presented to the eye; and therefore, if the officers who have charge of railway signals, that are red or green, should happen to have this imperfection of vision, most serious accidents might be the consequence.’
This was just one example of how those in scientific and medical circles felt about railway companies having colour-blind individuals as engine drivers or guards. Indeed, many of the individuals who spoke and wrote on this matter, especially in the 1850s when it became a widely discussed topic, called for the railway companies to introduce testing. In 1859 The Sheffield & Rotherham Independent detailed a lecture given by Edward Smith to a local Literary and Philosophical Society on the topic. He stated that “it was important that persons before they were made signal men, drivers guards etc. should be subjected to an examination, to see that they were not labouring under this defect.”
But, by this point some railway companies had started to introduce colour-blindness examinations. Smith cited the case of the Great Northern Railway who had instituted one for drivers, and by 1866 the Great North of Scotland had done the same.  Furthermore, in January 1880 J.H. Stafford, Secretary of the Lancashire and Yorkshire Railway, wrote to The Times in response to a letter that stated that many companies did not have a test for colour-blindness. He stated that his company had had a test in place for many years and that ‘no person was appointed to any position connected with the working of the trains’ without being tested. Indeed, it seems that by 1892 most companies had instituted tests for colour blindness for drivers, firemen, signalmen and guards. Indeed, by this point the London and South Western Railway even tested all individuals in the Traffic Department, who were not in safety-critical positions, and there is evidence that this practice was quite widespread.
Yet, a range of different types of testing methods for colour-blindness existed within the industry, and it is clear that some companies did not employ tests that were deemed effective by some in scientific circles. The Glasgow Herald reported that at a meeting at the Glasgow Philosphical society in March 1879 Dr J.R. Wolfe proclaimed that ‘the examination of railway applicants for application into railway service…he considers was worse than useless.’ Furthermore, at the Society of Arts in 1890 Mr Bond, the Great Western Railway’s surgeon, stated that the company’s directors had asked him to look into the matter. He agreed with the speaker, Mr Brundell Carter, who had said that ‘complaints had been made of the tests adopted by railway companies.’ Subsequently, by the 1890s there was seemingly a storm raging over the testing methods employed by railway companies.
These fears were validated by Board of Trade’s committee of 1892 on ‘colour vision.’ This was formed to look into the scientific understanding colour blindness and to determine which form of examination was the most effective. Despite there being no conclusive evidence by this date that colour-blindness had caused any accidents, the committee concluded that ‘very few [railway companies] have an adequate system of testing’ and that ‘nearly all the methods employed are defective.’ After taking evidence from 500 witnesses, it recommended that tests be compulsory and that railway company’s used ‘Holmgreen’s test.’ This used the gradual removal of different coloured wool skeins from a larger selection that had a range of colours. As the skeins were removed the examiner would narrow down the extent of the examinee’s colour-blindness.
Despite defining which test for colour-blindness was most effective, the committee only had the power to make recommendations to the railway companies. As such, it is unknown how many were using the Holmgreen’s system by the turn of the century. Yet, it is important to note that it was the State, and not the railway companies, that had to step in to determine which method of testing was most accurate. After 40 years of trying the railway companies had not by themselves been able to institute an effective industry-wide testing method and had evolved their own individual ways of doing things that were imperfect.
 The Derby Mercury, September 9, 1857
 The Sheffield & Rotherham Independent, December 03, 1859; pg. 8
 The Aberdeen Journal , December 5, 1866
 The Times, Jan 02, 1880; pg. 5; Issue 29767
 House of Commons Parliamentary Papers [HCPP],1894 [C.7325] Colour vision. Circular from the Board of Trade to the railway companies of the United Kingdom, and correspondence relative thereto.
 HCPP, 1892 [C.6688] Report of the Committee on Colour-Vision, p.17
 HCPP, 1894 [C.7325] Colour vision. Circular from the Board of Trade to the railway companies of the United Kingdom, and correspondence relative thereto.
 The Glasgow Herald, March 6, 1879
 The Times, Jan 23, 1890; pg. 11
 HCPP, 1892 [C.6688] Report of the Committee on Colour-Vision, p.17-19
 HCPP, 1892 [C.6688] Report of the Committee on Colour-Vision, p.20