History of Science Museum: Collection Database Search


Herschel's Early Fixing Experiments

Herschel's Early Fixing Experiments

In the first period of his photographic experiments (in 1839) Sir John Herschel's main concern and most notable (and famous) achievement was to discover and perfect a way of fixing the photographic image. There were other aims too (for Herschel's initial three-point agenda see narrative to 34848; for his ageing test see narrative to 94276), and there were various incidental experiments and discoveries - such as invisible writing (for example 78062), image reviving (43358), and trials in different atmospheric conditions (22719). But varied as they are in appearance and often in procedural details, the dominant theme of the experimental papers of 1839 is the fixing problem.

It was one of long standing. The darkening effect of light upon certain silver salts was known in the 18th century. The earliest systematic experiments with making images by means of this phenomenon, those of Thomas Wedgwood at the end of the century, verified by Sir Humphry Davy and published in 1802, concluded that 'Nothing but a method of preventing the unshaded parts of the delineations from being coloured by exposure to the day is wanting to render this process as useful as it is elegant'. It was by no means the only problem standing in the way of photography's viability; but it is surprising that no one followed up on Davy's conclusion (Wedgwood himself died shortly after) until Talbot and Daguerre, provisionally, and Herschel definitvely in 1839.

For Herschel in fact the solution was a foregone conclusion. He knew well the properties of hyposulphite of soda (now called sodium thiosulphate, but nicknamed 'hypo' by photographers), including its ability to dissolve silver salts, and had published a paper on them in 1819. It was consequently the first thing he tried, on the very day he started his photographic experiments, January 29, 1839, only four days after Talbot had first announced his invention of the photogenic drawing (without details). Talbot's fixative (and Daguerre's) was actually strong sodium chloride, though he tried other things with some success, especially potassium iodide; but there is no evidence he thought of hypo.

Herschel succeeded straight away. But of course there were many variables, both chemical and procedural, both in respect of the hypo itself (its strength, for instance) and in respect of the rest of the photograph (its initial chemistry, its exposure, even the type of paper it was on). Extensive experimental trials were required to find the optimum combination of the variables, and to perfect the procedure. These went on through February, and Herschel's preliminary results were announced on March 14, though his experiments continued until the autumn. The earliest dated image in the collection is February 7 (18828; see narrative to that entry). It illustrates both that hypo works, and that it is not yet perfect: stopping and preserving the image is one thing, but Herschel was keen to achieve perfectly clean highlights and strong dark areas and lines, without loss of fine detail (or such detail as the primitive photogenic drawing was capable of).

February's experiments include results more promising than that of February 7 (such as the February 26 oak leaf 60339), as well as failures (such as 85108, 89028). But then in a new series of experiments in June to August, concentrating now on combined treatments with hypo and water, the technique was brought to a high level of perfection (see narrative to 18871). The desired result was soon achieved consistently: an image of rich brown with perfectly clean light areas, good registration of detail, and resistence to subsequent fading and darkening (other perfected examples include 23691, 32610, 89167).

The difference between hypo and (say) sodium chloride is not just one of quality, they act quite differently. The latter is merely inhibiting further darkening, but leaving the volatile chemicals in place; whereas hypo, in Herschel's own perfectly chosen words, is 'washing out the superfluous silver'. In combination with thorough and even prolonged washing in water - though it took experience and experiment to realise this was needed - the hypo-fixed image was stabilised and the highlights cleaned, finally solving Wedgwood's long-standing problem. From 1839, hypo was the universally used fixative for silver-based photographic processes, and continued in use throughout the history of photography.

Meanwhile, Herschel naturally enough tried alternative fixing agents, especially during the earliest phase of his experiments in February. Those identified in his annotations on the actual experiments are: muriate of soda (sodium chloride, common salt solution, sometimes referred to in Herschel's notes as brine), ferrocyanate of potash (potassium ferrocyanide), hydriodate of potash (potassium iodide), hydriodate (an iodide, perhaps also potassium), hypo's sister chemical hyposulphite of ammonia (ammonium thiosulphate), and water (both distilled, and ordinary 'pump' water). The latter was sometimes effective, and water was also claimed as the fixative for Hunt's process (see 86717). Herschel's least satisfactory results, ironically, were with sodium chloride (favoured initially by Talbot). The others all produced some success, and are represented by surviving examples with discernible images in the collection; but none competed with hypo, and none had the known property of 'washing out the superfluous silver'.

The most interesting and promising results with alternative agents were afforded by the potassium salts, if only from the point of view of colour, which was something that greatly interested Herschel. Even the silver based processes, as Herschel's (and Talbot's, and Hunt's) experiments graphically demonstrate, were capable of a huge variety of colours, affected by various stages in the chemistry, not just at the fixing stage but in the initial (pre-sensitising) treatment of the papers too. The second period of Herschel's experiments (mostly in 1841-43) was devoted to the exploration and invention of alternative (non-silver) processes, characterised by highly distinctive colours - notably the blue cyanotype (see narrative to 81196), purple chrysotype (21518), and range of pastel-coloured dyes derived from flower petals (32274).

Potassium iodide under certain circumstances gave a pale yellow colour (see 31428 and 35834, unsuccessful but pale primrose yellow in February 1839; 46015 and 95542, slightly more success in June 1839). Talbot and Hunt both used it; and Hunt's direct-positive process, employing it as a sensitising agent in a bleaching process, gave results of a pleasant yellow and greenish-yellow colour (see 34450 and 62706).

While weak (or failed) treatments of potassium ferrocyanide might give nothing but the insipid fawn or pale sepia of many of the less successful experiments with hypo or salt, other perhaps stronger treatments resulted either (or both, sometimes) in a strong primrose yellow or a distinctive blue. Two specimens with both a yellow and a blue segment (16953 and 33336; and compare 60166) indicate Herschel's interest in these colour phenomena. They are in fact failed experimental images (Herschel's inscriptions and vestigial signs of a border show they were exposed contact copies of engravings). But the clear division into yellow and blue segments is not some accidental effect, any more than the over-all blue of 27352 (which is also divided into two segments, but in this case both are blue). Herschel has either treated the two parts of these papers with a different strength solution, or subjected them to some difference in procedure, specifically in order to explore the formation of these colours.

In 1841 or 1842 Herschel returned to potassium ferrocyanide and its close relatives, and other iron-based chemicals, for the second phase of his photographic experiments, in which he explored the photographic potential (and colour chemistry) of materials other than silver. This included the experiments by which he invented the cyanotype, or blueprint. The blue test sheet 27352 mentioned above might well, being undated, be thought a cyanotype experiment; but like its blue-and-yellow companions it definitely belongs to the early fixing experiments. The intial idea and inspiration for the cyanotype was thus derived from his observations of potassium ferrocyanide's behaviour and colour chemistry when trying it as a fixative in February 1839.

Other narratives:

Related Objects: