Month: July 2025

A Guimard’s metro signaling glass (“verrine”) for our museum project

( Note: in this article, the term “signaling glass” is used to translate the French “verrine”, meaning the protective cover of the lamp in a Guimard Metro candelabra)

 

If you could not attend our last Annual General Meeting, you may not have been able to see – or touch – our latest purchase: a Guimard metro candelabra signaling glass produced by the Cristallerie de Pantin. In a recent article, we corrected our earlier opinion [1] by admitting, with supporting evidence, that these signaling glasses were originally white, and that the change to an orange-red color took place around 1907, four years after Guimard’s collaboration with CMP (Compagnie du chemin de fer métropolitain de Paris) came to an end.

We would point out, however, that it was likely that red signaling glasses existed at an early date, since the detail of a black-and-white photograph of the uncovered surround of Rome station, taken in 1903 shortly after its installation, is more compatible with a red color than a white one.

 

On the contrary, white glass was used for at least one of the last surviving shield surrounds, installed at Porte d’Auteuil station in 1913.

 

At least one of these white signaling glasses still exists in a private collection, as we know it from a detailed photograph [2] taken in 1967.

 

A total of 103 open shield surrounds were installed between 1900 and 1913. One of these had already been dismantled in 1908, leaving 102 on the eve of the First World War. At a rate of two signaling glasses per surround, there were therefore 204 signaling glasses on the network at that time. Subsequently, the number of signaling glasses fell considerably because of the many dismantling operations that took place until 1978, when all the Guimard entrances were finally classified. At that date, there were still 60 uncovered entourages with shields. Logically, in the interest of optimal material management, the removed signaling glasses of the surrounds should have been stored, but we don’t know what happened to them.

As for the remaining signaling glasses on the network, whether red or white, they were all replaced by synthetic equivalents, probably in the 70s [3]. The motivation for this exchange can easily be guessed. It wasn’t a question of protection against vandalism, which was still in its infancy [4], but simply a maintenance constraint. These rather heavy glass vessels had to be handled when a lamp had to be replaced. This rather delicate handling, repeated dozens of times, led to numerous accidents on the necks of the signaling glasses and sometimes to their destruction. What’s more, the Pantin crystal glassworks had probably stopped producing new pieces a long time ago, so RATP was no longer able to renew them. The RATP therefore decided to replace them with copies made of synthetic material, which were lighter and less fragile, but far less beautiful. In the process, it dismantled at least a hundred signaling glasses, which should also have been put into storage.

 

By 2000, however, R.A.T.P. no longer owned a single one. What had happened? Unfortunately, the certainty that these signaling glasses would no longer be used on the network meant that this stock was probably managed in a less than rigorous manner. The lack of interest in, or even disdain for, the Art Nouveau style for many years meant that they had neither the artistic nor the financial value that would have encouraged R.A.T.P.’s management to preserve them. At the same time, however, private collections were being made, either out of aesthetic interest and without the feeling of committing a criminal act, or out of greed, since as early as the 1980s, Guimard metro pieces were being smuggled out of France, notably to make bronze copies for sale in the United States [5]. Authentic signaling glasses were found on at least one bronze subway surround in Houston.

Fortunately, before it no longer owned any, R.A.T.P. had, this time officially, loaned or donated signaling glasses along with complete surrounds, notably to the Museum of Modern Art in New York in 1958, the Staatliches Museum für Angewandte Kunst in Munich in 1960, the Musée National d’Art Moderne in Paris in 1961 (transferred to the Musée d’Orsay) and then to the Montreal metro company in 1966. During the restoration of the latter, the STP subway company also disassembled its signaling glasses, giving one back to the RATP in 2003 [6].

 

We concluded our previous article on the color of signaling glasses by prophesying that such glasses would inevitably reappear as the generations of their private owners changed. And this is precisely what happened: in October 2024, a few months after the publication of our article on glassware colors, one of our faithful correspondents — half-serious, half-amused — alerted us to the publication on a well-known free classified ads site of a proposal to sell the end of a subway candelabra with its glassware.

 

We immediately got in touch with the advertiser, who confirmed that it was cast iron (not bronze) and that the signaling glass was indeed made of glass. Given the rarity and interest of such an object, we quickly agreed on a price with the seller, and a family friend immediately dismantled it and put it in a safe place until it could be shipped to the Paris region.

 

As we had expected, the discussion with the vendor revealed that its owner (who had recently passed away) had held a fairly senior position in the RATP maintenance hierarchy, and that upon his retirement in the mid-80s, he had been given an entire metro lamp post. However, the size and weight of such an object made it very difficult to handle and use, so the happy new owner decided to saw off the end for outdoor use on his holiday home in the Pyrénées-Atlantiques region, where he was originally from. The remaining part rusted away for a few years outdoors in the Paris region, before being sold for its weight in metal.

Once the end of the candelabra had been recovered, we separated it from its sheet metal support. The signaling glasses had previously been removed from its housing. To do this, it was necessary to remove the cross-pin (held in place by a chain) which held the collar in place at the back.

 

The collar, held at the front by a hinge, can then be tilted to release the signaling glasses.

 

 

The post of the candelabra has been cut away, revealing the hole through which the lamp’s power supply passes.

 

The signaling glass was simply cleaned with soapy water, pending a more thorough cleaning. As we feared, its neck shows a lot of missing parts due to handling during lamp changes.

 

Weighing 6.5 kg, it measures 40 cm in overall length and 24 cm in height. Its maximum width is 26 cm, less than it should be (28 cm) due to the missing collar. The opening measures 26 cm by 20 cm.

 

Nevertheless, most of the signaling glass is in excellent condition and, with the difference that its neck is more uneven, our glass is identical to that of the RATP: same clean lines, satin-finish surface and color that varies according to the lighting and thickness of the glass, from dark red to light orange.

 

A piece of old glass sent to us by the seller shows that the glass is colored in the mass and not plated on the surface.

 

The fact that the signaling glasses were made in a crystal factory and not in a glass factory raised a question: was the material crystal or indeed glass? The shard above gave us an easy answer. Its weight is 8 g and its volume (obtained by dipping it into a graduated tube and measuring the rise in water level) is 3 ml, giving a density of 8/3 = 2.6, that of glass (that of crystal being 3.85).

 

Let’s recall the manufacturing process for this signaling glass as carried out by the Cristallerie de Pantin. A bivalve mold is used, articulated around its sagittal axis and based on a plaster model supplied by Guimard. To obtain the correct “point”, the glassmaker places a pellet of molten glass at the bottom of the mold. The gob of glass (the volume of glass picked from the jar at the end of the cane) is shaped by balancing and shaping, then introduced into the mold. A layer of about 8 mm of glass is then pressed onto the inner surface of the mold by air blown into the cane. The mold is then opened and the glass vessel cut with scissors to expose the wide opening. The edges are bent with pliers and probably shaped by applying another mold around the opening. After cooling, any imperfections and seams caused by the mold joints are carefully ground away. Finally, the outer surface of the signaling glass is acid-etched.

 

The aspect of this signaling glass has not failed to elicit comparisons with well-known shapes: teardrop, fruit, frog’s eye. One of the most disparaging comparisons, “half-sucked candy” [7], is not the most inaccurate. It’s not impossible that Guimard’s illustrative intent was to evoke a flame, as if spat from the end of candelabras. But now that we know that the first signaling glasses were white, this hypothesis seems less credible. It also seems possible that Guimard wanted to capture the look and workings of molten matter, with the signaling glass appearing to be both blown and then pinched at the end and stretched (this action being reflected in the wide striations around the neck). The idea of evoking a downward flow of viscous matter is also admissible, as Guimard was able to illustrate it, notably on the low-end posts of the secondary surrounds. As is often the case in his art of semi-abstract drawing and modeling, many interpretations are relevant, and everyone is free to formulate their own.

The Cercle Guimard was delighted to acquire this candelabra end piece and its signaling glass. It will of course be one of the centerpieces of the section devoted to the Paris metro in our museum project at the Hôtel Mezzara.

 

Frédéric Descouturelle

 

Notes

[1] Descouturelle, Mignard, Rodriguez, Le Métropolitain de Guimard, éditions Somogy, 2003; Descouturelle, Mignard, Rodriguez, Guimard L’Art nouveau du métro, éditions La Vie du Rail, 2012.

[2] As announced in our previous article, we will one day devote a special article to the astonishing batch of photographs of which it is a part.

[3] We give this very approximate date as a matter of conjecture. When writing the books on Guimard’s metro, we were unable to discover the exact date of this replacement.

[4] Unfortunately, at present, the reinstallation of signaling glasses, which  are easy and very expensive targets, seems to us to be completely unrealistic.

[5] See our articles on this subject: “The epidemic of fake bronze subway surrounds in the United States”: part one; the epidemic…part two; The epidemic…part three.

[6] The other signaling glass was entrusted to the Montreal Museum of Fine Arts.

[7] Quoted without reference by R.-H. Guerrand in Mémoires du métro, 1961. The article or book from which this quotation is taken has not yet been found.

 

Translation: Alan Bryden

A rare crystal vase by Guimard

22 May, 2025

Tracked for years, this exceptional vase has recently been added to the collections of our partner for the Guimard Museum project. Lengthy negotiations led to the acquisition of this rare object, the first of its kind to be identified and described. Unlike other areas of decorative arts where Guimard was prolific, he produced very little glassware.

 

The overall shape is fairly common for Art Nouveau glassware, resembling a “bottle vase” that can be divided into three parts: a very flat base highlighted by a protrusion, a slightly conical body, and a narrowing at the top before the opening curves outward. This shape, with its swirling ribs, seems to be unique to Guimard, who did not attempt to reproduce the silhouettes of his designs for bronze or ceramics. But as we shall see later, are we sure that Guimard is the creator of this shape?

This vase remained in the seller’s family for a long time, but it has not been possible to ascertain the original conditions of purchase. It is in good condition, despite a few scratches and a slight limescale deposit showing that it was once used as a flower vase.

It was produced at the Cristallerie de Pantin[1], as indicated by the signature engraved on its base, around the “STV” logo, which stands for the initials of the crystal factory’s full name: Stumpf, Touvier, Viollet & Cie.

 

Measuring 40.2 cm high and weighing 1.437 kg, it has a volume of 0.480 liters. These last two figures can be used to calculate its density: 2.99, very close to that of classic lead crystal (3.1).

 

The question of the exact nature of the material arose, even though the vase came from a crystal factory, because we established that the glass shades of the candelabra in the subway entrances produced by the same factory from 1901 onwards were made of glass without any added lead. The Cercle Guimard recently purchased one copy.

 

The base of the vase also bears Guimard’s handwritten, curved signature. It is characterized by the continuity of the design in a single line joining the first and last names and ending with a long initial that returns emphasizing them. It should be noted that the execution at the tip is slightly more hesitant than that of the crystal factory logo, for which the worker assigned to this repetitive task had a stencil.

 

In the list of contributors included in the brochure accompanying the series of postcards, mention is made of the Cristallerie de Pantin, located at 66 rue d’Hauteville, where their Paris store was located.

 

The crystal factory was included in this list of collaborators, certainly for the glass shades of the metro entrance that framed the entrance to Guimard’s pavilion, and possibly also for another product on display: probably vases, as suggested by the description that Guimard associated with the supplier’s name: “crystalware for decoration and furnishings.” Unfortunately, the views of the interior of the pavilion that we have do not show these vases. Moreover, there is no indication that the crystalware presented in the pavilion at that time was already the result of a collaboration between the architect and the Cristallerie de Pantin. Guimard may simply have chosen products from the glassworks’ catalog to decorate the furniture on display. The following year, however, we became certain that this collaboration had indeed taken place. In the booklet that Guimard published for the 1904 Salon d’Automne, he detailed his shipment by category. Number 4 of the “Objets d’art Style Guimard” is described as follows: “Guimard-style vase in aquamarine crystal. Price: 65 francs.“

The name ”aquamarine” was not invented by Guimard, who instead incorporated it into a production line of the Cristallerie de Pantin that had started a few years earlier. It refers, of course, to the name of the gemstone, a variety of beryl with a light blue color reminiscent of the sea. These bluish shades, which are difficult to achieve, were sought after empirically by various French crystal manufacturers. Émile Gallé in Nancy, for example, used cobalt monoxide to obtain a very light blue color called “moonlight,” which was unveiled at the 1878 World’s Fair. At the Cristallerie de Pantin, to obtain a slightly opaque glass with intermediate turquoise iridescence, the yellow of uranium dioxide and the green of a chromium oxide were mixed with cobalt blue.

The first mention of the production of “aquamarine” crystals by the Cristallerie de Pantin that we know of appeared on a page of the Maison Moderne catalog[2] published in 1901.

 

In the two photographs on this page, twenty-two vases of different shapes and a colors that can only be described as bluish all feature swirling ribs, often regularly punctuated with reliefs like the surfaces of shells, another way of recalling the marine inspiration of the series. The creator or creators of these modern shapes are not mentioned, suggesting that they were probably employees of the crystal factory[3].

Guimard’s crystal vase was not completely unknown to us, as documents from the Adeline Oppenheim-Guimard donation kept at the Bibliothèque des Arts Décoratifs identify at least two examples decorating a piece of furniture probably presented at the Salon d’Automne.

 

Within the same collection of documents, another twisted vase model, with a more pear-shaped base and a more tapered neck, is very similar to No. 4357 on the page of the Maison Moderne catalog (see above). However, Guimard’s involvement in this piece has not been confirmed.

 

Another example of a vase that could have come from the Cristallerie de Pantin is the translucent crystal model displayed in the showcase of the grand salon at the Hôtel Guimard, which features the same type of twists.

 

It should be noted that none of these vases feature any added floral decoration imitating the style of Gallé and Daum glassware from Nancy. This may have been a specific request from Maison Moderne, which wanted a more understated design in keeping with the Belgian style championed by Van de Velde. Outside of Maison Moderne, Cristallerie de Pantin produced numerous models of “aquamarine” vases with floral decoration, which occasionally appear on the art market.

 

Like these vases, Guimard’s vase also features an acid-etched decoration, but one that is unique to him and reflects his efforts to renew his semi-abstract style, gradually moving away from the harmonious arrangements of curved or whimsical lines that characterized his work around 1900.

 

Once again, it is not easy to describe or interpret this design, which leaves the observer’s imagination free to find its own references. While a plant-inspired motif seems to be ever-present, with highly stylized stems and flowers repeated ten times around the circumference, a motif inspired by the treble clef is not impossible.

 

Some of the undated watercolor drawings of frieze designs donated by his widow Adeline Oppenheim to the Musée des Arts Décoratifs show some stylistic similarities.

 

Other elements of the decoration, such as the blue lines and the yellow coloring at the top, allow us to speculate on how the vase was made. As with any glassmaking process, the glassmaker would have taken a parison from the furnace using his hollow rod and given it an initial shape by blowing and swaying it. He then rolls the parison on the “marble” (a polished cast iron table) to add calcined bone powder and arsenates, which have been placed there beforehand and which will give the glass a satin appearance over the entire height of the vessel and a yellow coloration in its upper quarter. In addition, twisted blue threads, visible at the top and bottom of the vase, were prepared in advance, laid down and then covered with a new layer of glass, which was returned to the furnace. This is therefore an intercalated decoration and not the usual Venetian thread technique, where the threads are applied to the surface.

 

With the glass vessel still at the end of the rod (on the side of the future opening), the base is roughly shaped by gravity and tapping against the marble. It is then placed in the clamshell mold and blown so that it sticks to the walls. This is when most of the swirling relief is created. Once the mold is opened, the vessel is picked up under the base using a small amount of molten glass at the end of an iron rod (the pontil). On the opposite side, the neck is twisted (as evidenced by the arrangement of the air bubbles that follow this movement) before being detached from the rod by cutting it with scissors and shaping it with pliers.

 

After cooling slightly to ensure stability, the vessel is coated with a thin layer of a more intense blue glass, which the glassmaker takes from a second furnace. The vase is then detached from the pontil, which leaves a scar under the base. After gradual cooling, the blue glass coating on the surface is treated with hydrofluoric acid. To reveal the decoration that will be preserved in blue, the surface is first protected with a varnish (at that time, Judea bitumen), while the unprotected surfaces are removed with acid until the underlying glass appears. Under the base, the trace of the pontil is removed by grinding and polishing so that the crystal manufacturer’s logo and Guimard’s signature can be engraved, as we saw above. However, around the opening, the remaining surface irregularities show that the glass has not been reworked by grinding.

 

The main reason why this vase came to the attention of Guimard specialists so late was its rarity. Its imperfect execution, particularly evident in its fairly substantial bubbling, suggests that it was part of a very small series. The other reason is that its attribution to Guimard is not obvious at first glance. It is even likely that, in the absence of a signature[4] and despite its abstract naturalistic decoration, this vase would not have aroused the same interest on our part, nor would it have prompted the research that followed to find the rare clues that enabled us to locate it within Guimard’s works. At the same time as he wanted to simplify the Guimard style by reducing it to a simple decoration on an object that could ultimately be adapted to all styles of interior design, it is quite likely that Guimard also chose a vase shape that already existed at the Cristallerie de Pantin — a shape that was not necessarily already on the market — and “Guimardized” it, thereby replicating the same approach he took with products from certain other manufacturers, as we have demonstrated for the stoneware vases produced by Gilardoni & Brault[5].

 

Frédéric Descouturelle and Olivier Pons,

with the assistance of Justine Posalski and Georges Barbier-Ludwig.

 

Notes

[1] The crystal factory was founded in 1847 at 84 Rue de Paris in Pantin and taken over in 1859 by E. Monot, a former worker at the Lyon crystal factory. It quickly prospered, becoming the fifth largest crystal factory in France and then, after the 1870 war (and the transfer of the Saint-Louis crystal factory to German territory), the third largest (after Baccarat and Clichy). The crystal factory industrially manufactured a complete range of products, which it displayed in its Paris store on Rue de Paradis. Its warehouse was also in Paris, at 66 Rue d’Hauteville. The company name changed several times: “Monot et Cie,” then “Monot Père et Fils et Stumpf,” and finally “Stumpf, Touvier, Viollet & Cie” after Monot withdrew in 1889. It was absorbed in 1919 by the Legras glassworks (Saint-Denis and Pantin Quatre-Chemins).

[2] Founded by art critic Julius Meier-Graefe at the end of 1899, this gallery dedicated to modern decorative art competed with the better-known L’Art Nouveau gallery owned by Siegfried Bing. Two articles by Bertrand Mothes will be published on this subject shortly.

[3] For other products, the catalog lists the names of the designers.

In this case, it is likely that the contract between the gallery and the designer specified that the latter’s name had to be mentioned.

[4] Another example of this vase has been catalogued and appears in the book “Le génie verrier de l’Europe” (G. Cappa, ed. Mardaga, 1998). Unlike our vase, which is predominantly aquamarine in color, this one is characterized by colorless crystal with a gradient of metallic tones, with the “rosathea” and “yellow” shades predominating, while the aquamarine tone is concentrated in the lower third. As this piece is unsigned, it may be a prototype.

[5] See our article on the “guimardization” of stoneware vases produced by Gilardoni & Brault.

 

Translation: Alan Bryden