for colour- ing to this theory, the mixed sensations of red, green, and blindness violet produce the sensation of white light; but evidently in induced by the cases where colour is absent in every part of the spectrum disease by the Young-Helm- except in the blue-the rest being seen as white-some holtz theory. different explanation is required. Or again, if we take into account the fact that at a certain distance from the centre of the retina all sensation of colour, varying according to its luminosity and its hue, is lost, though light is still seen, the ordinary application of the theory cannot be insisted upon.

Hering's theory and colour-blind

ness induced by disease.

Shift of the

neutral point

in the spectrum caused

by different qualities of white light.

It may seem that Hering's theory is fully capable of explaining most of these phenomena, but there are facts against its acceptance which are very weighty. For instance, according to this theory, the sensations of red and green, and of yellow and blue, ought always to be present together, but in some cases of colour-blindness caused by over-use of tobacco, and atrophy of the optic nerve, the blue is the only colour sensation felt, the yellow being absent from that part of the spectrum in which it should be present. Again, when the intensity of the light producing the spectrum is reduced the sensation of red disappears long before that of green, which shows that the two sensations are not always co-existent. The shortened spectrum of what are called the red-blind is also opposed to the theory, for the luminosity of the green is proportionally much greater to them than the red than it is to the green-blind.

NOTE (c).

The neutral point of the spectrum will vary in all cases of colour-blindness according to the whiteness of the light with which the spectrum is compared. Even to the normal eye there is a ray near the yellow which can match very closely indeed the light of a gas lamp or candle, though there is none which matches the whiteness of ordinary day- or sun-light. Now a match made by the normal eye of a coloured light with some ray of the spectrum will be equally a match to the colour-blind of either type, since in both the colour and its match in the spectrum the same one sensation will be absent. It therefore follows that their neutral point, with a candle or oil lamp as a standard of whiteness, must be the same yellow ray, but to the red-blind this ray would appear greenish if compared with the white of day-light, and to the green-blind reddish. If the mental picture of white light were that of day-light, then evidently the green signal light would have to be much bluer to the colour-blind than to the normal eye, to prevent a confusion between it and their neutral colour than would have to be the case when lamp-light is the mental image of white light. In testing a large number of men by lamplight it was invariably found that its light was always called yellow or orange by the normal-eyed, and we may therefore suppose that the general idea of whiteness is derived from

day-light. As this is the case with the normal-eyed, it may be assumed that the same mental standard of whiteness would be adopted by the colour-blind.

NOTE (d).

In a

In discussing the most suitable colour of signals, the question Effect of fog of the possible alteration of hue by the interposition of fog on the colour between them and the observer must be taken into account. of signals. There are white fogs and yellow fogs, the difference between the two being chiefly in the size of the particles of water, dust, or soot which are to be found in them. In a white fog away from large towns the particles are chiefly water, but whilst the great majority must be large compared with the length of a wave of light, yet some will be present which are very much smaller. yellow fog the fine particles are much more largely present, and the yellowness is largely due to this fact, for when particles, whose sizes are comparable to a wave-length of light, are present between the source of light and the observer, the law of scattering requires that the blue part of the spectrum of the light reaching the latter should be much more enfeebled than the green, the green than the yellow, and the yellow than the red. A blue-green signal glass will therefore appear rather less blue in a white fog, and even yellowish-green in a yellow fog, and it may happen that the loss of what are blue and green to the normal eye will shift the colour of the signal to the red side of the neutral point in the spectrum of each type of a colour-blind person, and then both red and green signals will appear of the same tint to him, though the latter will appear more diluted with the neutral colour. It follows therefore that in a fog the liability of the colour-blind to mis-read signals is very much greater than in ordinary clear weather.

EVIDENCE TAKEN BY THE COMMITTEE.

Evidence of Mr. HANBURY, of the Metropolitan Railway.

66

In the engine department, the men are examined as to perception of colour before they can qualify for drivers, but I think not for porters. If there is any doubt, we examine those engaged in traffic matters again, but not unless. We examine with the wool test, which I have here. We place this (a horizontal bar, from which were suspended skeins of wool about fifteen inches in length, and all bright colours) on the table in front of the man to be examined, and also a few skeins of wool, as an independent test. We ask the man what he understands by a danger signal; he says "red," naturally; then I ask him what colour represents a caution signal; he says green." I say, can you find the colour representing the danger signal. He looks, and perhaps picks out red; if he hesitates at all in his first choice, we ask him if he is quite sure it represents the danger signal. He perhaps says it does. Then we ask as to the caution signal; also test him with regard to the skeins of wool, and request him to pair or match the colour with a similar one on the frame; and if there is any doubt, we ask him as to brown or blue. Suppose he were to take this (mauve), we should test him again. I have not found many such cases on the Metropolitan Railway. Men sometimes mis-name the colours. We do not ask him the names of colours, but ask him to match them. We also ask him to pick out the " danger or "caution" signal colour, and we sometimes ask for the best red. We allow the man examined to make a minute examination between the colours. I cannot tell exactly how many men we have personally examined in this way, but I started my examinations in 1869, and have perhaps met with three cases of colourblindness. I cannot give an estimate as to the number examined. The wool test is the first test which my Chief undertakes, but when going on the footplate (on the engine) I examine them again myself. Agricultural labourers as a rule answer the questions as to the colour of the signals correctly. I never heard of the engine drivers rejecting the firemen, nor the case of a man going colour-blind subsequently.

Question. Do you have a certain proportion of men overrunning the danger signal in a way which cannot be accounted for?-I know of the case of a man at King's Cross passing the danger signal-an aged man-but I found his colour-sight good. The positions of the coloured wools on the bar are not shifted. The firemen are only tested once or twice as to colour, and afterwards if promoted to be drivers. We do not test with lamps, nor as to alteration of colour by fogs. We test them with regard to other colours than red and green if there is an

evidence of the necessity. We explain that the red signal is a danger signal, but we very seldom find a man ignorant of this, they generally know something of the work. There is no test with lamps, because the glass which gives the green light is blue by day-light. It is not a signal green glass, it is a blue glass [the glass is peacock blue]. We test by day-light. If we have green glass with the lamp we find it an indifferent light. The glass we use with gas is of a very definite green.

In case of hesitation, would you ask the man for some further examination, such as to pick out wool which was not far from a given colour?-We should not pass them if there was any doubt. If a man chose the wrong colour I should not think of passing him. It rests with the examiner and not with a doctor to pass a man. Cases have occurred where men have not been able to pass the examiner's sight test and have been sent to the doctor, who has given a certificate. We have a sight test; the test is with the single eye-one being covered; also with both eyes uncovered. We do not test our men at night as well as by day-light, further than already explained.

If accidents only happen at night, should they not be tested in that respect?-No, I think not. We take it for granted if a man can tell red in day he can at night.

The CHAIRMAN: Suppose that there were wools here, none of which matched that (red) exactly, but some nearly; if you were to ask a man to pick out a near match, and he showed hesitation, would you regard him as suspicious?-If he picked out the nearest, I should consider he had answered correctly. The picking out of an exact match does not prove that a man is not colour-blind. I have seen a man pick out brown, and call it red. Red represents the danger signal, green represents caution. On the Metropolitan Railway our signals are so arranged that in the event of any breakage of the glass the white light is treated as a danger signal. [Mr. Rix was here called, and the witness applied the different tests to him, the questions being answered to the satisfaction of Mr. Hanbury, who remarked that Mr. Rix had good sight. He was, however, informed that that gentleman was colour-blind. Mr. Hanbury stated that they would pass him on the Metropolitan Railway. Mr. Rix was recalled, and his colour-blindness proved by Dr. Grossmann's test.]

The CHAIRMAN: Do you ever use Holmgren's test?—I have never seen it.

Mr. BRUDENELL CARTER: If Mr. Rix were confronted with a single light he would not be able to tell which was green. His deliberation shows he cannot do it in a moment?-I must admit that Mr Rix being colour-blind is an eye-opener. We have about 500 men engaged in machine work, or on the engines.

Mr. BRUDENELL CARTER: Could you let me test them at some time by arrangement; it would not take long?—Yes, I should be very pleased to. I do not think painters painting various colours on the carriages and other things make mistakes

in colour.

If they did, it would be discovered during their

apprenticeship.

(By Dr. POLE): I do not know of any other railway using blue instead of green glass. I do know whether they use pure green. I have remarked that some glasses are bluer than others. We call them a better green. They are blue in daylight, but not by night.

(The Witness then withdrew.)

Evidence of Mr. WADDEN, of the London and South Western Railway.

The men entering the service of our Company are tested when they first enter, and again when they are promoted to be firemen, and every second year after that; and if during one of these biennial periods a man is promoted to driver, he is specially tested then. In the traffic department, every man is tested upon entering the service. They are tested in this way. I have brought the material in actual use for the purpose. These wools [the wools consisted of browns, drabs, sombre greens, one brighter green, and nondescript colours of very low tone] are placed upon a horizontal rod promiscuously, and the man asked to pick out three or four reds, blues, or greens, and if he makes an error in one of these, he is tested again with other colours. A man may have a good notion of colours, and not know what to call them. We do not find they mistake red for green. In addition to this test for the traffic department, in the locomotive department there is a night test. The room is darkened, and a box is fitted with a lamp at the back, and various coloured glasses are put in front, commencing with a small disc, perhaps the size of a pin's head, and gradually increasing till we get to one the size of a sixpence, the man being asked what colour he thinks is being shown to him. He is ten or twelve feet from the lantern. The smallest disc is the size of a pin's head, about one-eighth of an inch, or hardly as much, perhaps. We find the men are not so ready with the night test: they are more accustomed to colours by daylight, and find it easier in daylight to distinguish the colours than at night. These colours (wools) were provided by our storekeeper. I am not prepared to say under whose instruction. Mr. BRUDENELL CARTER: Among these there is not a single red. I should say they were selected by a colour-blind person! Capt. ABNEY: I am not colour-blind, but I should not know what to call some of these.

The WITNESS: Our locomotive foreman says many men fail in green who do not in red. I am told some men looked at that (green) and called it red. A further test is sometimes tried by sending men to the Ophthalmic Hospital, where there is a doubt, and I have been told that the hospital authorities confirmed our examination. This wool test is what we call the daylight test,