STAR COLOURS : 1

I N T R O D U C T I O N

General problems observing star colours is an interesting subject. As an astronomical subject, the attempts to explain the true cause of star colours originates dates back to the early to mid-19th Century. Much was thought and written by European astronomical circles, and this inspired the main visual observers of the times to investigate both single and double stars. The latter proved more unusual, as having two star close together allowed direct colour comparisons, which visually seemed to be far more colour enhanced. Perhaps, they had concluded, knowledge of these colours might lead to better understanding of the nature of stars — an aspect that continues now and likely into the future.

Beginning in the early 1800s, existed various conventional observations with an odd collection of accepted scientific or possible likely theories about the nature of star colours. Many then were popularly adopted by many visual observers, which continued with great interest and real furore into the early 20th Century. Without having the important advantages of modern astronomical spectroscopy or colour filter photometry, little was known about the evolution of stars, their intrinsic luminosities, nor even knowing the critical relationships between luminosity, stellar surface temperatures and all the observed stellar colours. There was also no broad-based knowledge of the true nature of radiation or the true power source that makes all stars shine so brightly nor of their ages.

Yet, by using some very careful and astute observations, some everyday useful information was obtained. They soon quickly learnt about general star colours by visual means were greatly lacking — mainly by the simple failure of human eye to function sufficiently well to see agreeable star colours in the nighttime. Compared to modern telescopes of today, many of them were of poor to very poor optical quality, often adopting difficult and cumbersome telescope mounts. This was made more apparent with the need for larger reflecting telescopes, whose basic available technology did included inadequately reflective surfaces from their metal speculum mirrors or significant light loss through optical glass in their often more crudely made optics or unsophisticated narrow-field uncoated eyepieces. These real observational restrictions all combined to make star colours observations as very subjective, error-prone and often quite flawed. In the end, from even their best-adopted experimentation and observational methods, they failed to reach any consensus or repeatability in their visual observations.

Until they had more objective means of producing empirical based evidence, their aspirations to understand the nature of stars from visual star colours remained simply unobtainable. The final death knell of naked-eye estimations of star colours was the discovery of astronomical spectroscopy, followed by using colour filter photometry, and then finding the basic scientific explanation of the electromagnetic nature of light and in how our eyes see colours. Today, much of this early work is mostly either irrelevant or unimportant knowledge without any positive contribution to observational astronomy or astrophysics. Sadly, even today, there still continue to be many poor and inaccurate views about star colours. Some of these antiquated notions and ideas have now continued to persist beyond a whole century.

COLOURS : DESCRIPTORS & ILLUSION

Another truly dreadful aspect of star colours in astronomy, which continues even now, has been the adoption of using far too many grandiose colour names. In some cases, this has literally expanding possibly many thousands of different colour names, technically known as colour descriptors or just descriptors.

One has only to look in your local paint shop and their colour charts to see the range of possibilities for their names. In truth, you can significantly increase the subtleties by mixing two or more coloured paints in different proportions to make new colours, and that does not include adding touches of black or white paint to make the paint darker or lighter.

In the everyday world, colour has vibrancy and shows a variety of optical phenomenon. Each day we see uncountable differences with colours, which we have learnt to take for granted — just as some normal everyday experience of life. No one seriously could be bothered to name them all.

Worst still, my own eyes likely do not exactly see the same viewed colour as you or even another person. It is like some lonely personal experience. Hence, my own colour descriptor may not exactly match yours. Like children, we combat this simple problem by describing the everyday life basic colour names like red, blue, yellow, green, orange, etc. If I see something that is blue, for example, you quickly know what my word means even though it might be different between you and me. Were the colour important, then I might use other associated words like dull, pale, light, strong, bright, etc., therefore extending the range to perhaps several hundred colour name variations.

Everyday experiences under daylight or indoor lighting finds even more diversity with colours. By changing the background lighting on any painted coloured surface; say from light sources that daylight, incandescent, halogen or fluorescent, and the perceived colour are significantly again different.^† Technically, this is related to the source colour temperature. For example, colours also change when viewing them on your computer screen display, being made by calibrating or setting the average daylight colour temperature to 6500K (actually 6503.6K) or so-called standard illuminant D65 define by 0the ‘Commission Internationale de l’éclairage’ / ‘International Commission on Illumination’ (CIE) via the available computer software preferences. Here temperature corresponds to the appearance of natural daylight by changing the display’s white point of the computer screen. Changing this temperature make higher values bluer and lesser ones redder. So importantly, colours appear differently when you compare them to subdued or bright surroundings.

^† Note: You can test this yourself using the website of paint manufacturer, Dulux, and using there Java software program called My Colour 4 and just follow the prompts. Here you can download an exampled image or add one of your own.

More controversially, colours may also affect your general mood, where reds and oranges are often heart warming, yellows are perceived as more happy and cheerful, while greens and blues can make you peaceful, cool or even calming. Dark colours often are more sombre in mood, while bright colours can be cheerful and pleasant. Contrasting colours can give dramatic effects like modifying the sense of a space, or highlighting some internal or external feature. Changing the background illuminating light source may make these effects even more dramatic.

If colours are so greatly experienced in all its diversity, then
why cannot we just simply apply these same rules to stars?

Well yes, you could, but it would not be very practical. The problem is simple to realise, as if you go outside in the dark at night, nearly all colour disappears. We see the wide gambit of millions of colours in the bright day, but at night, we mostly only see black, greys and white. Logically this is something to do with eye sensitivity under hugely varied illuminations. Our conclusion reaches to the idea that specific colours are not so obvious in our surrounds or the many faint stars in the nighttime sky. To argue otherwise is simply against commonsense. As an open question;

This broad-based article attempts to answer this question,
and lets you make your own general conclusions.

Reporting Nighttime Star Colours

It still remains a little hard to comprehend how such commonly expressed border-line or subjective colours can exist in stars. Ones like gold, crimson, lilac, indigo, grey or ashy can not be readily or usefully employed to describe star colours for most practical visual observers.

Some, perhaps, do often just innocently exaggerate the colours that they see. Probably they are only wanting to present these more exotic colours to make them seem either more original or accepted within the amateur astronomical community. Few may be nonetheless viewed as faux pas. Yet many do still continue to appear in articles throughout the popular magazines or in the observational astronomical press.

In my own humble opinion, several of them can only be described as new-age charlatans. Several wilfully claim in having either some kind of personal superior colour vision or special knowledge based on the quite whimsical notion of the observer being the better sex or having the better colour perception. I have even seen published star colours seriously presented as apricot, peach, amber, silver-white, lemon-brown, beige, khaki, or even turquoise! These were even mixed with so-called reflectance terms, like gloss, translucency or in illogical words like shadowy — whose the latter meaning is very uncertain here.

Why describe such insubstantial or unreal colours? All usefully observed star colours are far more straight forward!

In the end, such meaningless descriptions are just pure and utter nonsense because they are pseudo-visual colours, which verbally have quite arbitrary meanings — useless meanings that convey nothing at all to another person and are only useful to the individual that gave them! Worst with these kinds of observers is that the colours they are describing are physiologically impossible to see at night.

My reasoning of those persons stating these views is that they are presenting descriptive colours of so-called highly rich and saturated colours — something we will discuss and very much argue against in depth in this article. Furthermore, these importantly are also odd palette-like mixtures with the additional tones of black, greys or white — something that is not seen in the continuous spectra being observed with stars.

I really do think these types of amateur charlatans as observers must be immediately discredited, if only for the reason that they give very poor representations of the many good, sensible and dedicated amateur astronomers throughout the world.

Perhaps, as some earlier readers of this text have also openly said to me, That I am perhaps being a bit too critical of the situation. My sole aim and open wish here to highlight that using more specific or simpler colours schemes are far more sensible than in trying to match precisely what subtle shade of colouration one particular star or double star system appears to be.

Difficulties in Seeing Star Colours at Night

With star colours, much of the biological and chemical mechanism regarding colour vision unfortunately does not work very well at low illuminations. This is a major limitation for visual observers to overcome. These serious flaws really lie with the specialised cells known as cones located along the retina of the eye, being the main sensor that gains nearly all of the light needed for interpreting colour. It seems the human eye for all its true biological wonder was just never designed for good night vision. This is bad news for the amateur astronomer who is trying to perceive fainter objects and to see colour or spectral-based phenomena. Worst, there is no doubt that the age of the observer is likely another contributing cause for the eventual loss of the ability to interpret the spectral range. More unfortunate is that the younger the individual, the less able they can describe the visual colours they see just through lack of experience! Yet the real experts in recent times about eye colour perception have been made by several French visual observers, with several interesting papers in the last twenty to thirty years or so. For example, I have presented in Southern Astronomical Delights the translated version by Paul Biaze’s “Les Couleurs des Étoiles” or [The Colours of Double Stars] written in the late-1950s which is quite analytical and very innovative. A further excellent summary of this subject about star colour appears in David Malin’s Colours of the Galaxies (1996), which is recommended reading for all amateur observers.

Overall, the study of colour perception for stars is still incomplete. This general article is about the cause of colours that we see in telescopes and why they are so hard to observe. It was also written to counteract the seeming avalanche of several new double star observers who have been claiming that they have some superior vision or better colour perception.

Please, if you are one of those observers that believe what I am saying here is completely wrong, then I do suggest you reading the next four paragraphs very carefully before reading the rest of the text before you condemning me for evermore.

NATURE of EYESIGHT and COLOUR VISION

There is no known significant numerical differences of
rods or cones between human males and females.

During nighttime, visual observers find most star and deep-sky colours are mostly lost to our eyes. The simple reason is that cones have known thresholds for colour sensitivity, and below particular light energies (flux) they almost all completely cease to function. Consequently, when we look at our general surrounds during the night, we see only a slight range of “greyness.” Looking through any telescope, we are immediately exposed to the wide field illumination of the field stars and the astronomical object(s) in question. Most stars just appear white in colour, but in some circumstances, like the very blue or very red stars, we do begin to see some distinct colour. Also the fainter the star or object the less colour we see are able to discern. Hence, colour is also magnitude dependant. (Further discussed in Star Colours : 2)

Star colours that we see are quite different from what we mostly see during our everyday living because at night we perceive very few hues. This is due to the colour component known as saturation that can be described as the degree of whiteness in any perceived colour. Importantly, saturation is fairly weak in all stars. For many astronomical objects these will produce only pale or washed-out colours and never intense ones. The only true exception is probably the deep-red carbon stars which also visually appear to have a little blue or yellow light contributing to their general spectra and appearance. Such stars, however, are very unusual and rare.

Seeing star colours at night is unusual because
we can see no more than about 10% Saturation.

Experience finds that the more intense colours at night simply cannot be observed. The degree of saturation also only slightly varies between different individuals, and gets generally worst with age. Importantly it is also visually dependant on the background colour it is seen against.

Figure 1a. Variation of Colour Saturation

Shown here are the colours of red, orange, yellow, light blue and deep blue. Colour saturations above 10% are very rarely ever seen in stars or bright nebulae. 0% colour saturation is the pure white. All 100% saturation colours are often termed as pure colours.

Figure 1b. Variation of Star Colour Saturation

This shows an alternative view of colour saturation. The horizontal axis shows the variation in star colour from blue, through yellow and orange, ending in red. The vertical axis gives the percentage (%) colour saturations. As previously stated in the text, colour saturations at night are very rarely above c.10% seen in stars or nebulae. This is designated by the “ > < ” placed in the above graphic.

Important Note: How this figure appears to you will vary significantly depending on monitor quality and its adjusted calibration. It should NOT be used when compared to visual observations, as it is solely aimed just to highlight the importance saturation in stars at night.)

Figure 2. Effects of the Background on Visual Perceived Colour

This shows visual effects of 20% saturated colours as seen against either a black or white background. Each highlighted colour is identical, but visually our eyes see that those against the lighter backgrounds make the inside circle colour seem to be s lightly darker. This is visually caused by colour contrast interpreted by the eye, being comparable to looking at stars at night. For example, seeing stars during the hours of darkness when compared to seeing them against the background sky under either twilight or daylight. Similarly, double stars with quite different surface temperatures finds similar visual effects, which enhances apparent visual colour differences. Amateur observers should also note that increased magnification using different eyepieces makes a slightly darker background field and this leads to slightly changing the observed star colour.

Any real need for estimating the observed colour in telescopes is likely not very important for most visual observers. However, this is not absolutely true for those engaged in writing astronomical descriptions or in promoting astronomy. Such colour reports are both interesting and important to advise, whose knowledge may guide other double star or deep-sky observers towards some more attractive targets when observing.

How Much Reality is There In
Seeing Star Colours at Night?

Based on scientific optical experiments by visual physiologist Denis Baylor in 1978, it is now possible to conclusively dismiss misconceived notions of colour discrimination by telescopic observation. (See References) These original detailed experiments were conducted at the Department of Neurobiology at Stanford University whose main aims were specifically to measured the eye’s main photon response in darkness. Baylor attached a photoelectric photometer to individual rod and cone cells in human retinas, and then measured photoelectrically the response of the photons of various monochromatic colours. After analysing the results, his main conclusion found that at very low illumination, all cone cells switch off and cease their entire electrical function. This for the first time quantitatively explains the reason for loss of nighttime colour vision. Baylor further says about his general results;

“This state of affairs makes it impossible for one cell, either a rod or cone, to signal separately wavelength and intensity. Consider a single rod upon which falls 100 photons of 550nm. wavelength. These photons will be absorbed with the probability of say 10%, so that a total of ten absorptions will occur. Ten absorptions would also occur if 1000 photons were incident at 600nm. A particular wavelength therefore has the mean probability of absorption of only 10%. Since the cell reports only the number of photons absorbed, the signals generated by the two coloured lights are identical, even though their wavelengths are different. Hence no colour (wavelength) information is available. This explains why in starlight, where only the rods contribute to vision, we have no colour sensation.”

From this we can only conclude that as the rods receive the dim light, then our brains then try to interpret the actual colours it thinks it is seeing. Furthermore, as the star colours are never saturated, so what we generally see is only slight variations in hues. Hence, a narrow range of greys and only very bright saturated colours will be perceived as dull slightly coloured greys.

You can attempt such an experiment yourself. Look at a number of highly coloured book covers in the home under normal light illumination in the home. Turn off the lights, letting you eyes adapt for a moment, then look at the same coloured objects. If possible, turn on a more distant indirect light, and again observe the colours you see. In the end, you can see the reds, yellows and blues, but it becomes much harder to distinguish intermediate colours as readily under normally bright illumination.

STAR COLOUR SCHEMES

One of the first crude scientific star colour schemes was made by the variable star astronomer and editor of the Astronomical Journal, Seth C. Chandler (1846-1913) in 1901 (Chandler Scale — CI) using only seven basic colours to divide stars into simple colour groups. However, this idea was soon openly criticised, because it was so limited and unnecessary. Even the southern visual double star observer R.T.A. Innes (1861-1933) was one of the greatest critics of Chandler, stating that he placed little credence in knowing star colours as they could be equally obtained photographically using two colour sensitive films or by instrumentally by filter photometry. I could not find any related information in whether John Hagen did accessed the work of Chandler or Innes, but personally I do see much usefulness with the Chandler Scheme for most visual observers — mainly because it can easily distinguish these colours through the telescope. I’d also assume this would be exactly the same for the majority of people!

Innes seems to have reflected on this in his “Southern Reference Catalogue of Double Stars” (1899), but initially mostly ignored the colours of double stars. Those that he did mention colour were mainly the brightest pairs having clearly significant colour contrasts. However, he seems to have changed his mind between 1901 and 1903. In the updated and additional pairs of this catalogue in 1903 (“Micrometrical measurements of Double Stars 1849-1868 and 1899-1903”; Annals of the Royal Observatory, Cape of Good Hope Vol. II. Part IV.), Innes adopted a colour abbreviation system usefully describing double stars. This, I think, seems to have eventually became adopted into others like the Hagen Colour System. (See Below.) To reflect these changes, I have quoted his exact words, which Innes says in the “Preface” on star colours on pages. ix.-x.;

“…the fifth column gives the colours observed, wherein 0 will signify a white star, 10 an intense red star, and the intermediate numbers various stages from white to red, as follows” :—

It seems useless to go for further subdivisions or for fancy names of colour which can only convey distinct meanings to their actual author.

The close approach of Mars and γ Virginis, as seen with the naked eye in March 1903, affords a comparison. Mars would be 4 or 5 in the above scale, γ Virginis was decidedly bluish (b); this was due to contrast, as γ Virginis is a yellowish star.

Although the data as to colours are very incomplete (a) because if the companion is very faint it is impossible to estimate its colour, (b) because of a remarkable contrast in colour is unlikely to miss being recorded, the following rough analysis of the observations of colours of double stars is of some interest.

The general close accordance of the colour estimates given in the catalogue shows that such observations are not without value.”

This was soon superseded in 1924, when Rev. John G. Hagen (1847-1930) produced his new more logical colour scale. Hagen had specialised in eclipsing binaries, and also produced the famous Star Atlas called Atlas Stellarum Variabilium between 1899 and 1908. In time his name became synonymous with his colour scale that proved to be the more useful version of the previous and poorly adopted Chandler Index, simply known as the Hagen Colour Index (HCI), the scheme just labelled all star colours; ranging between the values of -3 for Blue and +10 for Red with neutral White corresponding to the B-V value of 0.0. This particular colour scheme has remained the popular nomenclature and is now often adopted by amateurs who do variable star observations or make micrometrical measurements of pairs.

Colours in this scheme were; Blue, Bluish, White, Yellowish, Yellow, Orange and Red. Hagen simply just adds additional colour values for these seven basic colour elements.

The HAGAN COLOUR INDEX (HCI)

Most visual observers tended to use the Hagen Colour Index (HCI) which relates closely to stellar surface temperatures and the B−V Colour Index. No one truly adapted this as an “analytical” method, but as an extra means of determining the “correct” position angle of both the stars, especially when the magnitudes are nearly equal.

Note: The original observer designations in double stars overrides the estimation of the brightest against the faintest star. This means the designation of A and B components are pre-set by the discoverer. The HCI probably has some analytical basis, however, the linearity with visual divisions in quite poor. I.e. The non-linear values, of say, white to yellowish stars are different, from say, blue to bluish or red to reddish stars.

Figure 3. The Hagen Colour Index (HCI) — 10% and 20% Saturation

This shows the Hagen Colour Index Scale with both 10% Saturation, the likely maximum visible colour, and 20% Saturation. I have contrasted the colours against both black or white backgrounds so the visibility of the colours and the contrast effects can be seen. The Figure above clearly shows these differences. All observed colours will be also slightly different when they are pinpoints, and the colour presented here are closer to the defocussed star images that can be seen in the telescope. Observers should note that calculated colours are approximately 10% and then I have had to make several small adjustments so that the colours look a bit more consistent. However, this changes are quite likely inconsequential for many visual observers. Most stars will be fainter than the colours presented here and nearly all of the fainter stars will have almost insignificant saturations.

Anyone using the colours for observational comparison should ONLY use the 10% SATURATION SCALE.

It was M. Minnaert who first discussed star colours in more modern terms. If we assume that star colours are based on the black body properties of objects, as seen in some ultra-hot furnace. A famous simple experiment of this is to continuously heat a small piece of tungsten wire in electric light filaments. Here the colours distinctly change as the temperature rises; from red-hot, yellow-hot, white-hot then blue-hot. This also similarly follows the observed spectral sequence of stars and the B−V colour index.

He then usefully adopted the series of eight separate colour groups as distinguish by eye. Then he did a simply blind experiment by comparing his colour estimates against the B−V colour index, which proved to have an observed high correlation. From this, he then first achieved the feat of being able to distinguish the spectral class letter of the object. Minnaert gained much kudos for this achievement in his day!

Minnaert also investigated the colour of the white and yellow stars, finding that they could distinguish the yellow ones into white-yellow, light yellow, pure yellow and deep yellow. (The reason for this, I think, is that the eye is more sensitive to seeing this part of the spectrum, especially when compared with the red, and the far blue.) Interestingly, his experiments validates the problems of colour saturation. His book concludes that only eight major or primary star colours each corresponding to the mid-spectral classes of O, B, A, F, G, K, M, S.

Figure 4. Colours of the Spectral Classification — 10% Saturation

Figure 4 shows the colours of the Spectral Classification at 10% Saturation. The colours can be estimated in the telescope with care, but observers should note that these are the maximum colours and most of the stars have much lower saturated colours. I have contrasted the colours against both black or white backgrounds so the visibility of the colours. The colours here are suitable for using in drawings of star charts where the spectral class is required.

Again, anyone using the colours for observational comparison should ONLY use this 10% SATURATION SCALE.

According to David Malin (AAO), it was the astronomer Leslie Morrison from the Royal Greenwich Observatory who attempted visual observation of stars through the transit telescope, and doing a blind test, could guess the Spectral Class of the star in question! Each class could be seen and ascertained with the eye, with each have only three or four shades of certain colours, with the solitary “non-colour” of white. There are fourteen “valid colours” in this second system, being in order of;

The MINNAERT COLOUR SCHEME

DESCRIPTORS FOR DOUBLE STAR COLOURS

For double star observers such methodologies have been already established using scales like the Hagen Colour Index (HCI). This scale has values between -3 and +10, describing the possible range of fourteen double star colours — from blue to white to yellow to orange to red. This roughly mimics the range seen in astronomical spectra, in stellar surface temperatures and spectral classes. However, the problems for double stars observers is that using this particular index, finds the fundamental inherent weakness is a scale is that it does not differentiate between the different colour saturations. Furthermore it takes no account of the stellar magnitude. Although this scale is quite arbitrary between observers, different eyes will certainly see different colours. Unfortunately, the HCI system leaves too large a range of observable possibilities for the many different colours. Moreover, detecting colour is also very observationally troublesome to see as the stars more often than not appear simply as point sources. Often by just simply defocussing the star into small plate-like disks can be applied to partially exacerbate this problem.

When reading some older books, texts and catalogues, you will sometimes find the use of the following abbreviations. (See Table below.)

These main colour can also have the following additions: Colour that are less bright than normal are prefixed p pale — or if brighter in colour are r rich or d deep. I.e. pale blue, rich yellow, or deep red, etc. Colour tendencies towards any colour are sh but is very rarely used.

Unconvincing colours, for example, a suspect yellow star would be Ysh (Yellowish) or Bsh (Bluish).

A further usefulness for this colour scheme is that the observer can quickly write down these abbreviations in his or her observation notes. Although the use of colour is likely not important, but it is an additional descriptor when checking the pair at some later date to differentiate equally bright components or in reducing observations.

Later use of the abbreviations now tend tofavour the Hagen Colour Index (HCI), which relates closely to stellar surface temperatures. Using this index, visual observers should report as e.g. “−2 / 3”, being its pale blue primary and pure yellow secondary. Other additional colours were added later I.e. −0.5 for grey and −0.25 for green.

REPORTING STAR COLOURS
GENERAL RULES

If the colour is a definite colour, report it as eg.“White” or “Blue” etc.
If the colour seems a definite tint, report it as eg. “Yellowish” or “Bluish” etc.
If it seems either like a combination or range of colours report it as “Bluish” or “Bluish-White” etc.
If the colour can not be described, record it as “Unusual” or “Colourless”
If the primary’s colour is seen but not the secondary, record it as “Blue / − ” etc.

For a continuance of this page: See Star Colours 2. (Next)

Last Update : 23rd May 2017

Southern Astronomical Delights © (2017)

For any problems with this Website or Document please e-mail me.