British photographic history

Information and discussion on all aspects of British photographic history

Does anyone know of a table that correlates the no. de fabrication of autochromes (stamped on the boxes) with the date of production? Or know anyone who might know?

(OK they are French, but they were used by the Brits ...)

Hopefully

Giles Hudson

Views: 464

Reply to This

Replies to This Discussion

Giles,
I have not seen a table which correlates the "no. de Fabrication" with a production date. Autochrome boxes usually have an expiration date stamped above the Fabrication number though often times it is very faint.
A method that can be used to date boxes should the expiration date no longer be visible is to see how the name Lumiere appears on the top of the box. A "Societe
Lumiere" nameplate will date it from before 1911. If the name appears as "Union Photographique Industrielle Lumiere & Jougla (U.P.I ) " then the box was manufactured after Lumiere & Jougla joined forces in 1911. Of course, there is no guarantee that exposed plates remained in the same box in which they were shipped.

You also might try contacting Bertrand Lavédrine and/or Jean-Paul Gandolfo. They recently published "L’autochrome Lumière, secrets d’atelier et défis industriels" which is by far and away the best book on the industrial process which produced the autochrome.
http://www.crcc.cnrs.fr/spip.php?article209&lang=en

Mark
Hi Mark

Thanks for those suggestions.

Unfortunately autochrome boxes in the early years (maybe pre 1910?) don't have the "a employer avant" dates on them. Certainly for fabrication numbers less than 150. After then the best before date ought to be a good indicator - allegedly they were only good for three months.

There ought to be other ways of dating autochromes in the absence of boxes: from the grain structure. After 1909 I believe they stopped filling the gaps between starch grains with carbon black, and I suspect there are other forensic changes over time. I would be interested to learn if anyone has ever investigated this.

I shall follow up on L’autochrome Lumière, secrets d’atelier et défis industriels. I wonder if anyone knows if there is a copy in a library in England?

Giles

Mark Jacobs said:
Giles,
I have not seen a table which correlates the "no. de Fabrication" with a production date. Autochrome boxes usually have an expiration date stamped above the Fabrication number though often times it is very faint.
A method that can be used to date boxes should the expiration date no longer be visible is to see how the name Lumiere appears on the top of the box. A "Societe
Lumiere" nameplate will date it from before 1911. If the name appears as "Union Photographique Industrielle Lumiere & Jougla (U.P.I ) " then the box was manufactured after Lumiere & Jougla joined forces in 1911. Of course, there is no guarantee that exposed plates remained in the same box in which they were shipped.

You also might try contacting Bertrand Lavédrine and/or Jean-Paul Gandolfo. They recently published "L’autochrome Lumière, secrets d’atelier et défis industriels" which is by far and away the best book on the industrial process which produced the autochrome.
http://www.crcc.cnrs.fr/spip.php?article209&lang=en

Mark
Giles,
On "date of utilization", Lumiere noted the following on a 1921 box:
" The Date of utilization printed on our boxes of Autochrome Plates give but a minimum indication, beyond which the plates may be used and will still yield good results during four or five months..."

Yes, one method that can be used to date a few early autochromes sans the box ( which is not terribly reliable in any case ) is by identification of the type of final varnish applied to the plate. This method has been successfully used to date some of Steichen's early autochromes.
Lumiere's original varnish was composed of a 20% solution of gum dammar in benzene which was to be poured over the plate. Two competing recipes soon appeared which added a second resin, either gum mastic or Manila copal. The effect of both these resins was to increase the hardness of the varnish and thus more scratch resistant. In 1908, Steichen experimented with Zaponlac, a cellulose nitrate varnish, which he described in Camera Work. This varnish offered certain advantages since it was insensitive to heat and dried quickly. However, it did have a tendency to lift the emulsion on some plates.

A full description of these variations and how they were analyzed can be found in the chapter Photography in Natural Colors: Steichen and the Autochrome Process by Tinia Passafiume, and Coatings on Autochrome Plates by Bertrand Lavedrine, Clara von Waldthausen, and Lyzanne Gann in Coatings on Photographs. Materials, Techniques, and Conservation. Ed. by Constance McCabe, 2005, AIC.

I do believe carbon black was used to fill-in the interstices between the grains after 1909. According to Sam Welford, the Charcoal specified was to be freshly pulverized in a ball mill, 3 kilos of charcoal in 10 litres if dist'd water. The mill was to run for eight days and the charcoal recovered in a vacuum filter and dried in an oven.
Charcoal fillings were lessened and then finally eliminated with the introduction of Filmcolor, Lumicolor and Alticolor since greater pressure could be applied to film than to glass.

As to Fabrication numbers, I have examples of earlier boxes with larger numbers and later boxes with smaller numbers. It would be quite interesting to learn of any correlation between those numbers and use date. Not sure how the plates which were manufactured in Burlington, Vermont would figure in on those dates though.
Interesting ...

What I was basing my assumptions on was Comte de Calmas “Using Old Autochrome Plates”: “MM. Lumière mark on the boxes of the Autochrome plates a date, which represents a period of about three months, later than which the plates are not advisedly used ...” (1 Apr 1910). But the question is really just: how many months after date of manufacture did they date the best before date? It would be nice to get a definitive answer to this. But again this is a value that may have changed over time.

In my small sample of about 20 boxes none with fabrication nos less than 262 have best before dates. This corresponds to mid 1909, I think. Which made me think 1 April 1909 might have been a change-over date - on the introduction of the simplified process.

On the filling:
“Starch-Grain Screen-Plates.—No. 8,153, 1911 (April 7, 1910)” Lumiere autochrome variant “In the new process, the mixture first projected is formed of coloured particles of all the colours except one which are to appear on the screen, the last colour being obtained by the projection of finer particles ... in the interstices left by the larger particles ...”
But perhaps this patent was never put into practice? I have not been able to examine any plates under a microscope.

On the varnish: I guess that is a good way of dating them if you have an x-ray flourescence spectrometer to hand! (which I do, but have never been able to convince the necessary curator to be brave enough to take advantage of it ... )

Have you any idea, by the way, why some autochromes go completely green? Not just green spots, but absolutely uniform green all over, obscuring all other colours? Is this a very long-term aging problem, that wouldn't have been present at the time they were made? I have not seen any mention to the uniform green fault in the pre-1912 literature.


Mark Jacobs said:
Giles,
On "date of utilization", Lumiere noted the following on a 1921 box:
" The Date of utilization printed on our boxes of Autochrome Plates give but a minimum indication, beyond which the plates may be used and will still yield good results during four or five months..."

Yes, one method that can be used to date a few early autochromes sans the box ( which is not terribly reliable in any case ) is by identification of the type of final varnish applied to the plate. This method has been successfully used to date some of Steichen's early autochromes.
Lumiere's original varnish was composed of a 20% solution of gum dammar in benzene which was to be poured over the plate. Two competing recipes soon appeared which added a second resin, either gum mastic or Manila copal. The effect of both these resins was to increase the hardness of the varnish and thus more scratch resistant. In 1908, Steichen experimented with Zaponlac, a cellulose nitrate varnish, which he described in Camera Work. This varnish offered certain advantages since it was insensitive to heat and dried quickly. However, it did have a tendency to lift the emulsion on some plates.

A full description of these variations and how they were analyzed can be found in the chapter Photography in Natural Colors: Steichen and the Autochrome Process by Tinia Passafiume, and Coatings on Autochrome Plates by Bertrand Lavedrine, Clara von Waldthausen, and Lyzanne Gann in Coatings on Photographs. Materials, Techniques, and Conservation. Ed. by Constance McCabe, 2005, AIC.

I do believe carbon black was used to fill-in the interstices between the grains after 1909. According to Sam Welford, the Charcoal specified was to be freshly pulverized in a ball mill, 3 kilos of charcoal in 10 litres if dist'd water. The mill was to run for eight days and the charcoal recovered in a vacuum filter and dried in an oven.
Charcoal fillings were lessened and then finally eliminated with the introduction of Filmcolor, Lumicolor and Alticolor since greater pressure could be applied to film than to glass.

As to Fabrication numbers, I have examples of earlier boxes with larger numbers and later boxes with smaller numbers. It would be quite interesting to learn of any correlation between those numbers and use date. Not sure how the plates which were manufactured in Burlington, Vermont would figure in on those dates though.
Giles,
On the green problem:
I too have some plates which are completely green. These plate were all done by the same maker. Are your green plates by various autochromists or by one maker?
The problem may have been caused by improper use of hypersensitized plates. Most autochrome instruction manuals have a section dealing with "Hypersensitizing Autochromes". The manual notes:

"When properly hypersensitized with our solution and correctly exposed with our Magnesium Screen, autochromes yield pictures free from any prevailing tint.
A predominance of yellow would show that that the immersion in hypersensitizing solution was too long or the solution too warm.
A predominance of blue would show the opposite defect.
A predominance of red or orange shows that the bath was too concentrated.
A predominance of green shows that it was too weak...

The green cast is also mentioned in the Colour Supplement of the British Journal Of Photography, Page 8, Feb 1, 1918. (Light-Flter for Hyper-Sensitised Autochromes)

Another factor which might have caused green plates, and this is pure speculation on my part, is that one of the after-treatments attacked elements in both the red and blue dyes while leaving elements of the green dye relatively intact. If this is true, then perhaps an explanation would be the chemical make-up of the final varnish. G. E Brown and Welborne Piper recommended celluloid dissolved in amyl acetate as opposed to the varnish formulas discussed earlier. However, according the E.J. Wall, "E. Koing strongly advised against this or any varnish containing nitrocellulose, as the latter under the action of light gives off nitrous fumes, which would be apt to attack the dyes of the screen elements." ( Wall, p.528 ) Alas, neither Wall nor Koning state if this attack was uniform against all the dyes or worked selectively against some while leaving others more or less intact or whether that formula perhaps acted in such a way as to essentially be a dye itself.
The green ones are all by the same photographer, but they are only a small subset, and appear unconnected with each other. The strange thing about them is that they are otherwise very fine, and have been bound up as normal, suggesting there was nothing wrong with them at the time they were made. They could have been varnished differently, but I doubt it. Could perhaps have just been a bad batch. They are probably all 1915 or later.

I know the green was allegedly the most soluble of the dyes, so I wondered if damp over several decades had just caused all of it to spread, obscuring the red and violet.

Their greenness has caused one curator I know to identify them as Agfa process rather than autochromes at all. I don't think I buy this though.

Giles

Maybe you have already contacted Bernard Lavédrine and Jean-Paul Gandolfo, just like I did many years ago when the Rijksmuseum had acquired some 250 autochromes made by Jacob Olie Jr. These were partly kept in Autochromes boxes. Some had dates, some not.

 

The ones that had noth a number and a date ('A employer avant') are:

 

1288: Fin Septembre 1914

1318 2: Fin Novembre 1914

1668-2: Fin Septembre 1922

1733: date unreadable

1786: Fin Novembre 1924

1949: Novembre 1927

 

Jean-Paul Gandolfo was then so kind to send  me a list that I think am allowed to give here:

 

331: Septembre 1909

350: Octobre 1909

383: 15 Janvier 1910

518: Novembre 1910

604: Juillet 1911

712: Novembre 1911

713: Octobre 1911

828: Juillet 1912

833: Juillet 1912

928: Octobre 1912

957: Decembre 1912

959: idem

1058: Aout 1913

1091: Septembre 1913

1113: Octobre 1913

1116: idem

1130: Novembre 1913

1238: Juillet 1914

1483: Février 1918

1487-2: 2 Mai 1918

1575-1: Novembre 1920

1618-1: Octobre 1921

1675-1: Novembre 1922

1710: Aout 1923

1718: Septembre 1926?

1744-2: Avril 1924

1775: Septembre 1924

1800-2: Avril 1925

1911: Decembre 1926

1934: Novembre 1923

1936: 1927

1947: Novembre 1927

2007-2: Aout 1929

2043-2: Aout 1930

2050: 1936

 

I hope this may help to establish dates and I guess that Lav''edrine and Gandolfo have by now much more numbers and dates.

 

Hans Rooseboom, Curator of Photography at the Rijksmuseum Amsterdam

Many thanks for this. I wonder if the number of plates per batch is know?

Giles

Thanks Hans!

 

Here's a few more:

 

181:        Aout 1908

1158:      Decembre 1913 

1207:      Mai 1914

1245:      Aout 1914

1314.1    Octob. 1914

1540:      Fevrier 1920

1579-2:   Fevrier 1921

1579-2:   Fevrier, 1921

1607:      October 192?

1661-2    ?           192?

2867:      Mai 1926

2867:      Mai 1926

2867:      Mai 1926

2867:      Mai 1926

2867:      Mai 1926

2961       Juin 1928

Reply to Discussion

RSS

© 2020   Created by Michael Pritchard.   Powered by

Badges  |  Report an Issue  |  Terms of Service