German English    

Reflections on the 1870 Iron Cross With special consideration to the Second Class

Mike Estelmann


Perhaps no other military decoration is as significant as the Iron Cross. At a minimum, this statement is true for the foundation year series of 1813 and, at a respectful distance, the 1870 series. The historical dimension, the engaging symmetrical aesthetic, and the appealing juxtaposition of two metals which could not be more different from one another, only begin to explain why contemporary examples of this patriotic decoration are in such high demand. As this demand is now so high, and there are an insufficient number of originals, there has followed what must: modern copies. These copies may be revealed through the use of relevant literature, photo documentation, and comparative analysis with known originals. To begin with, of course, we have the words of the Old Master of Prussian Militaria-writing. Louis Schneider, the Privy Court Councilor and Reader of his Majesty the King, notes in his 1872 work, The Book of the Iron Cross, that King Wilhelm I re-instituted the Iron Cross with an order dated 19 July 1870, the very day of his beloved mother Queen Luise's death. He continues: On 30 July, the day before the departure of the king to join his army, the president of the General Order Commission, Adjutant General v. Bonin, presented an 1870 Iron Cross, finished that same day, to the king. Wilhelm himself wrote on the accompanying letter from General v. Bonin: "Approved." Thus the order was given to the director of the Iron Foundry, Bergrath Schmidt, to begin casting the crosses according to the model. On 11 August, Director Schmidt announced to the jewelers authorized to make the silver frames that the Iron Cross cores could be picked up. Furthermore, Schneider recounts his manufacturing figures in detail, and states in summary that, by command of His Majesty the King, a total of 44,489 crosses had been delivered as of July, 1871 from the General Order Commission to the Military Cabinet. The foundry to which Schneider refers is the Royal Iron Foundry in Berlin, founded in 1804 and located in front of the Oranienburger Gate. In addition to the Royal Iron Foundry in Gleiwitz, the Royal Iron Foundry in Berlin manufactured cores for Iron Crosses during the Napoleonic Wars of Liberation. For our purposes here, we must investigate the manufacturing techniques and connections of the time. The Berlin Foundry had a high technical "know-how," and when undertaking to manufacture 1870 cores its artisans could look back with pride on a long and successful company history. From their workshops, for example, were built, in 1816, the first two gear-wheel steam locomotives in continental Europe, the Kreuzberg Monument for the Wars of Liberation in Berlin, and the Berlin castle bridge, which to this day testifies to the Foundry's efficacy from its spot on Berlin's Invalidenstraße. But the Foundry did not undertake only large-scale projects. Its experience with decorative artwork and delicate jewelry seemed to make it the perfect choice to manufacture the cores for the Iron Cross. Let us investigate the workshops and production techniques to learn more about our subject. After the confirmation of the new core design, the first step was to make a wax model from which a master mold would be made with a soft metal like tin or silver. Then sand -- sifted several times through a linen cloth and kneaded with clay -- was mixed with water and tapped by a hammer into the mold until it was tightly compacted. After smoothing, the workpieces were pressed in and then the second, already prepared side of the mold could be attached and assembled under high pressure. After the separation of the two halves -- which could be accomplished without difficulty as the contact faces had been powdered with coal dust before -- the master pieces could be removed leaving the remaining imprints connected by fine channels. This is how a so-called casting tree was produced. After the finished casting molds dried, the actual casting began. Experienced workers poured the molten iron from a melting pot into the provided openings of the mold, where the iron flowed evenly into the hollows. Through previously-applied outlets, the suppressed air escaped. This technique allowed the production of high quantities in the shortest time (Fig. 1).


Fig. 1: The illustrated magazine "Stein der Weisen," published at the beginning of World War I, shows the tradional process for manufacturing the Iron Cross. Initially, each core was cast. However, the astounsing number of Iron Crosses awarded -- over 5 miliion in World War I -- soon made other techniques necessary. 


After the solidification process of the iron, the molds were separated and the cores removed by means of a slight knock, or tap, on the side of the mold. With a polishing stone the cores were finished by hand and given the desired look of fine jewelry. Then followed the process of annealing, or removing inner stress through a process of heating and gradually cooling again. Next, the cores were again evenly heated, and a dark varnish made from linseed oil, resin, and galena black carbon was applied. A final rapid heating caused the oil to evaporate, leaving the iron cores with that durable matte-black finish typical of the Iron Cross. This finishing process is also known as "false blackening," and has nothing in common with real blackening in which there is a chemical reaction of the base material itself and a kind of patina forms. In the case of "false blackening," a thin, highly adhesive coating adheres to the core, creating the so-called "blackening." From the evidence so far observed, we may now draw our first conclusions. The majority of Iron Crosses awarded during the campaigns of 1870-1871 must have identical cores, since they were all manufactured at the Royal Iron Foundry in Berlin. Minor differences are due to the aforementioned manufacturing and finishing method. Only the frames should exhibit marked differences, since they were made by several different jewelers. But here again we must reference Schneider, who states in his book that as late as March 1872, several thousand Noncombatant First and Second class crosses remained to be awarded. He does not mention where these later-awarded examples originated. A further order from the Royal Iron Foundry is probable, and Schneider may have made no mention of it on the assumption that it was understood. With the following photo documentation we may try to further understand our subject. It should be noted that an assignation of a manufacturer to the 1870 EK2 can only come from comparison with marked First class crosses(Fig. 2). 


Fig. 2: Iron Cross First Class marked for Johann Wagner & Sohn, Berlin


Exactly this core type, with its distinctive date and crown design (Fig. 3), is to be found on most Second class Iron Crosses, and it may be assumed that it is this type described by Scheider and made in the Berlin Foundry. 


Fig. 3: Iron Cross Second Class. The core design is identical to the FIrst Class example illustrated above. 


Further clues are to be found in the available literature on the history of military decorations. Jorg Nimmergut, in Volume 2 of his book, German Medals and Decorations to 1945, shows an engraved EK1 from ex-Kaiser Wilhelm II's Huis Doorn, with a core identical to those previously described. This cross is his father Crown Prince Friedrich Wilhelm's personal example, awarded on 20 August 1870. Also, former Chancellor Bismarck's EK1, shown in the recently-published book of his awards, exhibits an identical core. Furthermore, this core type may be seen in contemporary photographs (Fig. 4-6). 


Fig. 4: The War Minister to-be, Karl von Einem gen. von Rothmaler. 


Fig. 5: Lieutenant General Bernhard of Austria


Fig. 6: Details of Iron Crosses worn by unknown recipients 


This core type shall now be designated "Type A." All examined Iron Crosses with this Type A core have the following characteristics: - The cores are cast. - The size varies by tenths of a millimeter, and only very rarely exceeds 42mm. - The weight can be between 15.6 and 17.5 grams. - The jump ring is affixed very near the top and is usually open (unsoldered) on one side. Never has a maker mark or a silver-content mark been seen on the ribbon ring. (Fig. 7) 


Fig. 7: Jump rings open (unsoldered) on one side. 


The cores are not painted, but blackened as described above (Fig. 8) 


Fig. 8: Reverse of a "blackened" Type A core. 


After core Type A, we must now consider a second type with different characteristics. Again, assignment of a maker has been accomplished by comparison to marked First class crosses. This core shall be called "Type B." The illustrations show a First class Iron Cross made by Godet, Berlin, and an example of a Second class with the same core (Fig. 9-10).


Fig. 9: EK1 marked for Godet. Characteristics of this core include the slim numbers in the date, the slightly leaning "8", and the tall, narrow "0". 


Fig. 10: EK2 with leaning "8" and high, slender "0". 


Supporting contemporary evidence is again provided by "The Great Nimmergut." There is a cross of this type in the possession of Kaiser and King Wilhelm I and shown in The Prussian Cultural Heritage Foundation. This core type is also easily discernable in contemporary photographs (Fig 11).


Fig. 11: The Type B core can clearly be seen on this unknown veteran's bar. Photo: Aron Willers, Friedrichshafen. 


Type B cores have the following characteristics: - The cores are cast. - The size varies by tenths of a millimeter, and occasionally exceeds 42mm. - The determination of weights is somewhat distorted by the fact that a number of examined crosses exhibited repairs. At the lower end of the scale is 16 grams. However, it is important to note that no example weighed over 17.6 grams. - The cores are not painted, but blackened (Fig. 12).


Fig. 12: Reverse of a "blackened" Type B core. 


- The jump ring is also attached near the top, is usually open on one side. Again, no manufacturers or silver-content marks were seen. The evidence above permits the conclusion that there were two contemporary core types, designated here as Type A and Type B. The essential characteristics -- casting, blackening instead of painting, high and almost identical jump ring, weight and size tolerance -- are very similar. Only in the design of the core details may differences be found. The continuing pursuit of our quarry leads us to the year 1874. Although not previously noted in the study of this subject, it is nevertheless true that the Royal Prussian Iron Foundry in Berlin was closed in this year. Kaiser Wilhelm I issued the order for closing the Foundry on 31 March 1873, and the last cast was made on 5 January 1874. The inventory of the foundry was sold at auction to other state enterprises and institutions. Not only does this development represent the end of a historically and culturally valuable era of art and Iron work in Berlin, it also raises the interesting question of who, after the Foundry's closing, was able to manufacture cores to meet the demand for replacement Iron Crosses. Inseparable from this question is the existence of core Type B. As we have seen, Type B cores were used by Godet of Berlin. Moreover, as of this writing, they have been seen exclusively in Godet crosses. There exist no other crosses clearly attributable to a different manufacturer with this core type. This stands in contrast to Type A cores, which are clearly to be found in crosses assembled by jewelers other than Wagner of Berlin. Let us remind ourselves of Schneider's observations: "Director Schmidt announced to the jewelers authorized to make the silver frames that the Iron Cross cores could be picked up." If you infer from this statement that several jewelers arrived to pick up the finished cores from the Berlin Foundry, it becomes clear why Type A cores are to be found enclosed within frames manufactured by multiple companies, and why there are EK1s marked by various jewelers, including Godet, with Type A cores (Fig. 13).


Fig. 13: EK1 with Type A core, but marked for Godet. Photo: Markus Bodeux, Herne.


But Type B cores are known only in Godet frames. We may now conclude that Godet either made, or had made, Type B cores. An interesting corollary may be found in secondary literature. In Friedhelm Heyde's standard Iron Cross reference book on the collection of Max Aurich, published in 1980, the author writes about recipients of the 1813 Iron Cross who received their awards after the end of the Napoleonic Wars: Whether the casting of the iron cores was accomplished in the Kgl. Preuß. Eisengießerei Berlin (Gleiwitz) or at the iron foundry of the respected manufacturer of religious jewelry, Godet, has not been conclusively established. This means, of course, that Godet did have the means at their disposal to make the iron cores themselves. Also, as anyone interested in the history of military decorations is aware, there was a strong urge for companies to keep everything in-house -- both manufacturing and design. Examples of Godet's own urge to individuate their designs are to be found in their unique swords, the completely different design of their First class Prussian Red Cross medal, and their stylistically divergent Prussian Stars. Core Type B, with its completely different design, fits in this list rather nicely. Such a venture would have been economically beneficial as well, for not only were duplicate pieces needed, but replacement crosses would have been required by those who either lost their originals, or whose originals suffered from the common breakage of the jump ring. The latter problem was already known from the first crosses. Why the problem was not fixed for later crosses is not known. Perhaps, as has been suggested, the research and retooling required was not cost-effective, an example of questionable craftsman logic. (Fig. 14-15). 


Fig. 14: Clearly visible is the repair to this EK2. 


Fig. 15: A period repair to the jump ring. 


At this point we may make a preliminary summary of our findings: Core Type A and Type B are both contemporary to the award period. Type A cores, made at the Berlin Foundry, were distributed to several jewelers, but predominantly Wagner of Berlin, for use in award and private-purchase Iron Crosses. Type B cores, probably manufactured in-house by Godet, were used in duplicates and replacement crosses. In this context, it is worth mentioning that crosses with core Type B are found far more frequently mounted on medalbars (großen Ordensspangen) than are Type A crosses, a fact which tends to support the theory that they were manufactured as secondary pieces. After this excursion into manufacturing, production techniques and historical context, we must now describe yet another type of cross. It is a type which can not be found in any known period groupings, nor can it's provenance be established by any other supporting documentation or evidence. Moreover, it is only in recent years that this type has been seen in the marketplace. This statement is further confirmed by the fact that no examples of this type may be seen in Friedhelm Heyde's book on the Aurich collection. Neither can such a type be identified in Harald Geißler's 1995 Iron Cross book. In Jorg Nimmergut's previously cited book, published in 1997, there are only images of core Type A and core Type B crosses. Only newer publications, such as American author Steven Thomas Previtera's The Iron Time, published in 2007, show this type of cross. (Fig. 16-21) 


Fig. 16: The two crosses in the top row are clearly larger than the Type A and Type B core crosses in the bottom row. 


The characteristics of this type of cross are listed below: - The cores are not cast, but stamped. - The crosses are larger and vary in size from 42 to 44 millimeters. - They are much heavier and generally weigh over 19 grams. - The ninth bead in the headband of the obverse crown is generally somewhat larger than the others, slightly offset to the bottom, and quite noticeably protuberant. 


Fig. 17: Obverse crown with protuberant ninth bead. Also note maker's mark "Z" or "N" on the ribbon ring. 


- The crossing lines on the "8" on the core date are not on the same level as in Type A cores, but rather one crosses noticeably over the other. This is known as an "over-and-under 8." 


Fig. 18: The "over-and-under 8.". 


- On the reverse side date, there are deep flaws in the lower portions of the "8" and the middle "1".


Fig. 19: The date flaws on the reverse. 


- The jump ring is heavily soldered and sits lower on the frame; the ribbon ring almost always exhibits a marker's mark. 


Fig. 20: Maker's mark "L" on the ribbon ring, and the characteristic ninth bead in the crown's headband.


- The cores are all painted and not blackened.


Fig. 21: Clearly visible is the core's paint, and the oversize ninth bead. 


That these crosses are not originals should be evident by Schneider's remarks as well as by the total absence of illustrations and discussions in earlier literature. However, the argument that these are original pieces, or contemporary duplicates, is heard time and again. There is no doubt that there was a demand for replacement crosses. Indeed, the need to address breakage problems and the 25th anniversary of the Iron Cross's re-institution would have raised this demand. Thus it is not proper to dismiss outright any cross that does not have either a Type A or a Type B core, as the market for second pieces surely gave rise to different variants. Figures 22 and 23 show an example that was certainly a contemporary secondary piece. The painted core of this example resembles a Type A, but clearly deviates in some details. 


Fig. 22: Obverse of a replacement cross. 


Fig. 23: Reverse of a replacement cross. 


However, in light of known manufacturing techniques, the cross is question shows itself to be highly suspicious. Size and weight of these pieces correspond to the average values of World War I Iron Crosses and are significantly over accepted tolerances for core Type A and core Type B 1870 Iron Crosses. The jump ring attachment is identical to 1914 EK2s, and the stamped cores are painted, again as with 1914 EKs. But the list of suspicions is not yet exhausted. Maker's marks are found on the ribbon rings. These markings are easily identifiable as World War I codings. At this point, the weight of the evidence clearly establishes that these crosses are modern counterfeits, or fakes, made from newly minted cores and genuine World War I Iron Cross frames. This fake is called "The Ninth Bead Fake," and is known with the following maker's marks on the ring: L, WS, Wilm, N or Z, KO, CD 800, MFH, G, K.A.G., L.W., IVI, R-W -- next to KO, K.A.G. and CD, the most common maker's marks from 1914-series Iron Crosses. The instinctive notion that these may be legitimate duplicates made for veterans during World War I may be rejected on two counts: first, the logic of biology dictates that demand would have been very low for 1870 Iron Crosses after 1914, and second, no contemporary manufacturer could have made the quantity of these Ninth Bead Fakes on the market today and still met their obligations to manufacture 1914-series crosses; they dominate today's market in disconcerting numbers. It would be a mistake, however, to imagine that there was no production of 1870 Iron Crosses after 1918. There was indeed a small, but verifiable, production of such awards. The manufacturing quality of Ninth Bead Fakes, however, does not compare favorably with World War I-made pieces. This discrepancy, combined with their increased numbers in recent years, does not permit any classification other than contemporary fake. This verdict is further supported by the extremely unprofessional way the frame halves have been rejoined after the cores were exchanged. Original pieces always exhibit a finely soldered seam -- testimony to the high skill of contemporary silversmiths. Mastery of this skill may be confidently assigned to World War I-era craftsmen also; thus the Ninth Bead Fakes can not have been made during either period. Moreover there is not even the slightest possibility that the frames were opened and rejoined during World War I, as anyone with a need would have had access to freshly made original frames (Fig. 24).


Fig. 24: A selection of poorly resoldered frames on examples of the "Ninth Bead Fake."


This last unambiguous evidence that the Ninth Bead examples are fakes -- the unprofessional resoldering of the frames -- may signal the final disappearance of these fakes from the hobby and the market. But the final farewell may not have yet been heard, unfortunately, for it must be mentioned that, although rare, First Class examples of this fake have been seen. 


Special Thanks to: Trevor, Markus Bodeux, Herne, Michael Fischer, Ladenburg and an anonymous specialist. 


Literatur: Arndt, L. / Müller-Wusterwitz. N., Die Orden und Ehrenzeichen des Reichskanzlers Fürst Otto von Bismarck, Offenbach, 2008. Geißler, H., Das Eiserne Kreuz, Norderstedt, 1995. Hessenthal, W. v. / Schreiber, G., Die tragbaren Ehrenzeichen des Deutschen Reiches, Berlin 1940. Heyde, Friedhelm, Monographien zur Numismatik und Ordenskunde, Preußen-Sammlung Max Aurich, Teil C, Das Eiserne Kreuz, Osnabrück, 1980. Meyers großes Konversationslexikon, sechste Auflage, Leipzig und Wien, Bibliographisches Institut, 1905. Nimmergut, J., Deutsche Orden und Ehrenzeichen, Band 2, München, 1997. Previtera, S.T., The Iron Time, Richmond, 2007. Schneider, L., Das Buch vom Eisernen Kreuze, Berlin 1872 Schreiter, Ch. / Pyritz, A., Berliner Eisen, Hannover, 2007


© Andreas M. Schulze Ising VII/2009

© 2005