The common coins of the mid-4th century

Expanded to include Horsemen, Galleys, Huts and other types
This WWW site began in February 1997 with a piece on a minor variety of a coin of Constantius II. While my page on Stirrups? is still available, it seems time to expand the discussion of this period more generally. These coins were introduced as part of the coinage reform of 348 AD by Constantius II and Constans. All coins of this reform bear the reverse legend FEL TEMP REPARATIO ('Restoration of Happy Times') which was translated by Victor Failmezger in his excellent survey of the issues (The Celator, Vol. 6 No. 10, October 1992) as 'Happy Days are Here Again'. The scope of this page and my knowledge of the subject will not allow coverage in minute detail. I feel particularly weak in my understanding of the denominations involved with this series. The earliest FEL TEMP REPARATIO coins were clearly divided into three denominations with Falling Horsemen and the Galleys being the largest. The middle denomination was almost the same diameter as the largest but contained less silver and was distinguished by left facing busts on their obverses. These commonly show the reverse type of a soldier leading a barbarian from a hut or the emperor with two captives. The third denomination was appropriately smaller and bore the popular design showing a Phoenix. As time progressed the smaller two denominations were discontinued and the Falling Horsemen shrank in a series of standards revisions to a size smaller than the smallest 'Phoenix'. The exact relationship of each of these coins to each other and to the gold coins of the day is a subject being studied as this is being written. One sees explanations of these coins that apply denomination names 'Majorina' and 'Centenionalis' but I remain unconvinced that we are correct in the application of these names to the coins in hand. For now, at least, I prefer to use the size scale AE1 (over 25mm), AE2 (21-25mm), AE3 (17-21mm) and AE4 (under 17mm). Earliest FTR coins are AE2 while the last issues had shrunk to the border between AE3 and AE4.

Part 1 - The Falling Horseman

Five assorted barbarians
On the Falling Horseman type, Virtus or a Roman soldier is shown spearing a barbarian horseman. Varieties show this barbarian in different poses but in all he is dying. Of all Roman coins showing a battle scene, this series is unusual in showing the moment of death and contends with a Republican denarius (of M. Sergius Sileus) showing a severed head as the most violent and gruesome Roman coin. The barbarians who died on your coins were from several different tribes and countries. Some can be identified with some degree of certainty by their attire, hair and beards. This study is very much in progress as this is being written. Unfortunately, many coins are crude and leave room to intrepretation of what is being shown in the small details. The number of different enemies shown differs according to the significance a student places on these details and whether they result from the intent of the cutter or are simply artifacts of crude workmanship. A comparison of historical descriptions of the enemies of Rome and what we see on coins should prove extremely interesting in this regard.

By the time of these coins, only two sons of Constantine the Great survived and had divided the Empire under an uneasy truce. Mints tended to issue more coins for the brother that controlled that region but at least some coins were struck for each at every mint. The Falling Horseman was a favorite type of Constantius II. Falling Horsemen were struck for Constans at mints controlled by Constantius II but his own mints preferred his equivalent coin type, the Emperor on a galley sailed by Victory. Civil wars involving the usurper Magnentius and the brothers resulted in the death of Constans in 350 AD and made the Falling Horseman the favored coin type for the next decade. Magnentius issued a small number of Falling Horsemen at the beginning of his reign but soon switched to is own distinctive reverses. Falling Horsemen also were issued by Constantius Gallus and Julian as Caesars appointed by Constantius II but when Julian survived and succeeded Constantius II the type was retired. There are no Falling Horsemen of Julian as Augustus.

Late Roman Bronze Coinage(LRBC) by Carson, Hill and Kent recognize four design groups of Falling Horsemen based on the pose of the barbarian. These are:

FH1 Kneeling Horseman kneeling on ground before horse
FH2 Sitting Horseman sitting on ground before horse
FH3 Reaching Horseman on horse turning and reaching back toward the soldier
FH4 Clutching Horseman on horse slumping forward clutching horse's neck

SLG* Lugdunum mint workshop 2
Horseman Kneeling - * following mintmark
TSA Thessalonica mint workshop 1
Horseman Sitting - A in fields on both sides
ASIRM Sirmium mint workshop 1
Horseman Reaching - A behind bust / III in field
ANQ Antioch mint workshop 9
Horseman Clutching - G in reverse field

LRBC goes a step further and separates (as FH2a) coins of the sitting variety that have a star in the field but stops short of recognizing other (more significant, IMHO) differences within the types. RIC does not number the types but describes them more completely under each mint. For this page, we will use LRBC numbers augmented with differences described by RIC . Fifteen mints scattered about the Empire produced Falling Horsemen. No single mint produced all four varieties but each of the fifteen struck some FH3 'Reaching' coins (by far the most common variety).

TSA* Thessalonica workshop 1
Bust right with globe
LRBC 1647 - RIC 116 page 412 vol. VIII
While most certainly handy, the LRBC division of these coins into FH1 through FH4 is not without problems. RIC (Vol. VIII, page 38) recognizes only three types combining 'Kneeling' and 'Clutching' and placing the arrangement in a different order. Certainly separating FH-1 and FH-4 coins can be difficult in some cases. When the horse is standing up and the barbarian is on his back with arms wrapped around his neck, the coin is definitely FH-4. To be FH-1 'Kneeling' the barbarian should be on the ground in front of the horse. In most of these questionable coins, the horse lies prone on the ground making it less clear if the rider is on the horse or on the ground. The photo in this paragraph illustrates such a coin. I consider two additional points in assigning this coin. First, FH-1 horses tend to lie with their heads stretched forward rather than curled under (forehead to the ground) as seen on FH-2 through FH-4 coins. Also, the soldier on earlier types (FH-1 and FH-2) tends to be shown on top of the horse more than the later pose where he stands behind it often with one knee on the rump. Using these criteria, this example would seem FH-1 but LRBC lists it as FH-4. Is this correct or just a typographical error? Wear and opinions can cause problems in such minor differences. More grievous is the error on LRBC plate II swapping the photos for 2625 and 2295 confusing the definitions of FH-3 and FH-4. It is good to see that I am not the only one to have errors on my pages. If you find one of mine, write and I can correct it. That is what I love about web publishing. :)

FH-2 'Sitting' also shows some variations in details. Our example above shows the barbarian fully seated on the ground. Other coins show him in a half seated, half standing pose as if he were either struggling to his feet or falling backwards using one leg to break his fall. Both are considered 'Sitting' types. Of the poses, the common FH-3 'Reaching' is the most consistent. The horseman turns around and reaches one arm back to his attacker. In some he appears to be fending off the spear but most depict a futile gesture.

The thirteen years that were spanned by the Falling Horseman type saw many changes (up and down, mostly down!) in size and weight standards. The early coins were what collectors term AE2 measuring 21 to 23mm diameter; the final issues were down to about 16mm not really qualified to be called AE3. Often these later, tiny coins are called AE3/4 recognizing that they are still a bit large to term AE4. All were silvered when first issued but relatively few retain more than a trace of the very thin coating. Workmanship declined with the size. Die quality varies greatly within the series; some mints approached crude!. The Falling Horsemen can be big, bold and beautiful or simply horrid little scraps of bronze. Various issues were distinguished by mint marks and control letters which would enable those 'in the know' to separate the current issue from last years coins which were called in and replaced by new coins of 'appropriate' weight and content. Silver content varies from about 3% to almost nothing.

The above photo shows ten assorted Falling Horsemen from the period 351-361 AD. Weights range from 9.5g (top row, #2 which is abnormally heavy) to 1.7g (a Julian Caesar resting on a U.S. cent for size comparison). Mint and series marks appear in the table below. All are Constantius II unless noted.

G reverse field
G reverse field
9.5g (thick)
G (dot) reverse field
(dot) S (dot) reverse field
D obverse field
G reverse field
B obverse field
4.2g (thick)
reversed Z after mintmark
(dot) reverse field
Julian II
M reverse field
Julian II
M reverse field

GSIS crescent Siscia mint - Horseman Reaching
A behind bust / "72 to the pound" in field
Scattered around this page are a number of examples showing just a few of the types and marks available. Some issues bore 'series' letters on obverse, reverse or both. Some series were distinguished by dots. One even bore the numeral LXXII attesting that the coins were struck 72 to the pound. The meanings of all of these 'codes' are not known with certainty but there are serious students working (and disagreeing, unless I miss my guess) on the details. This is one series I believe has realistic hopes of having a definitive text published in the next few years. This will be a great boon to collectors of common late Roman copper. No Falling Horseman coins are terribly expensive. 'Floor Sweepings' are under a dollar. Decent examples of the small, later ones can be found for $5 to $10. Perfect (REALLY Perfect!) examples of the big ones can push $100 but it is a very rare Falling Horseman that brings over $50 and 90% of all seen are worth under $20. This makes them a great coin for a student collector who wants to have several coins and seek out 'flyspeck' varieties. From the Falling Horseman beginning, a specialty collection could easily grow to include all the FEL TEMP REPARATIO types and, then, all Constantinian era coins. For the money, I consider these common coins to be the best bargain (interest and fun per dollar) available in ancient numismatics.

Postscript: Included on this page is one coin weighing 9.5g (Group photo, top row, second coin). Have others run across other examples of Falling Horsemen greatly over the weights recorded in RIC? I would appreciate hearing from collectors having greatly overweight FH coins or having opinions on the weight standards of the series.

Part 2 - The Galley

Just as Constantius II favored the Falling Horseman, Constans most frequently used the Galley steered by Victory. The scene probably honors Constans' crossing to Britain in 342 AD. The emperor is shown standing in the galley holding either a small figure of Victory or a Phoenix on a globe. While standards changed over the two to three years of issue, the short period does not show nearly the variations seen in the Falling Horsemen. There are, however, coins of half size bearing this type suggesting that some mints used the Galley both for the largest and smallest of the three denominations used in the early FEL TEMP REPARATIO coins. Our illustration shows a large (AE2) Constans with Victory (Trier mint, left) and a half size (AE3) Constans with the Phoenix (Thessalonica mint, second from left). Later FTR Galley coins included Magnentius (Trier mint, center) who used the type briefly before converting to his own distinctive types. This coin is smaller than the first Galley issues. I am uncertain where to place these medium size coins in the scheme of denominations. As mentioned above, the denominations of the FTR series needs more study. Some mints once under control of Constans also issued the type for Constantius II (Thessalonica mint, second from right). Most of these seem to date to the period after the death of Constans. After the death of Constans in 350 AD, the type was used for AE2 Galley coins in the name of Constantius Gallus (Siscia mint, right). Following these, the type was retired and replaced by Falling Horsemen. A similar design with different legend was revived during the era of Theodosius I.

Part 3 - The Barbarian and the Hut

The middle denomination of the FEL TEMP REPARATIO group was distinguished by the use of left facing bust portraits holding a globe in the hand. These AE2 coins were only slightly smaller than the largest denomination but contained less silver in the alloy and were probably valued at half the large Galley and Falling Horseman coins they accompanied. The reverse scene shows a soldier leading a small barbarian from a hut located under a 'tree'. The interesting point on these coins is that they were issued by 13 different mints and each used a different type tree (branch, shrub or plant). Perhaps botanists will be able to identify the species intended. Coins without mintmarks can be placed by the style of the tree and hut. The type was used more commonly for Constans but Constantius II versions also exist. Our illustration above shows an obverse of Constans (far left) and Constantius II (second coin) followed by reverses from six different mints. Like the Falling Horsemen, these coins often show field letters or stars that distinguish issues but, like the Galley type, the Huts were discontinued on the death of Constans so the variation in sizes is much more limited than with the Falling Horseman series.

Part 4 - Emperor & Two Captives

The middle denomination type favored by Constantius II shows the Emperor with two captives. Again there are minor variations but not nearly to the degree seen on the Hut coins. Our illustration shows two coins of Constantius II. On the left, Antioch mint uses two captives squatting on the ground while, on the right, Nicomedia placed the captives on their feet. Of the various FTR types, most collectors consider this one to be the least interesting. Even J. P. C. Kent writing the introduction to RIC Volume VIII (page 35) referred to them as "essentially banal". As a sidelight for those interested in minutia, the inset shows the officina letter from the left (Antioch) coin which appears to have been changed in the die from H(8) to Z (7). On modern coins, this would be a great discovery but few people collect ancients with such minor die oddities.

Part 5 - The Phoenix

Finally we consider the smallest denomination of the FTR series. Weighing half as much as the large (AE2) coins and containing only a trace (accidental?) of silver, these could not have been valued at over half the middle denomination. Obverses were right facing busts of either Constantius II or Constans but the type was discontinued before the first issues of the other rulers discussed above. The Phoenix coins were issued by ten mints in two major varieties. Six mints show the phoenix standing on a globe while five used a pile of stones (pyre). Thessalonica issued coins of both varieties. Our illustration shows reverses from Siscia (pyre, left) and Nicomedia (globe, right). Symbol of rebirth, the Phoenix type seems particularly appropriate for a series of coins issued on the 1100th anniversary of the founding of Rome.

Our little survey of the FTR coins is far from a complete reference on the subject. There are other varieties not mentioned here as well as hundreds of versions made special by rare combinations of field letters and mint marks. As this is being written I am aware of two separate efforts to write a book on the coins of this period. Although common, they are extremely varied and present a great opportunity for building an interesting collection at reasonable cost. Publication of a good book on the subject would be a great service to numismatic scholarship. I hope we will soon see the fruits of these efforts.

Return to main page

(c) 2000 Doug Smith