A vehicular bridge crosses the upper Alde at the east end of the Straits, White House Farm, Great Glemham (GR TM 356 619). Underneath the bridge there is a concrete ledge on each side of the river, raised somewhat above normal water level but submerged in flood. From time to time, on no particular routine, I check these ledges for evidence of mink or otter. Both have left their signs in past years. In early August (2007) it appeared that the exceptional high waters of July, which overflowed the river banks, had cast up onto the broader (eastern) ledge an old brick and a large bone (See below). On these prominent objects, including both ends and the shaft of the bone, mink had deposited their characteristic smelly scats. According to Corbet & Harris (1991 p. 406) this is normal behaviour, but it is interesting that these sites were well used so soon (within two weeks) after being exposed by the return to normal water levels.

The bone is a complete left femur of an equid, stained dark brown as from long immersion in water or mud (See second photo). Comparative measurements show that it is too large to be the femur of a donkey (a skeleton at Cambridge Zooarchaeological laboratory, kindly measured by Dr Phil Piper), but is shorter than the femora of two Arab horses in the Natural History Museum, London. Calculated from other bones, the heights of these Arabs at the withers were 1529 mm (female) and 1480 mm (male), i.e., roughly 15 and 14 hands, respectively (information kindly supplied by Dr Juliet Clutton-Brock).

In length and mid-shaft diameter the Alde femur does, however, fall within the dimensions of horses excavated at the Danebury Iron Age fort (Grant, 1984, 1991). Dr Annie Grant has commented that the most commonly used multiplication factor for calculating withers heights from horse femora is given in an 1888 publication by L. Keiswalter, referenced in A. von den Dreisch & J. Boessneck (1974: Kritische Ammerkungen zur Widerristhohenberechnung aus Langemassen vor- and frühgeschichtlicher Tierknochen). This method is based on the greatest lateral length/greatest length. For the Alde femur, GL indicates a horse about 125 cm at its withers or just over 12 hands. Another study of horse size quoted by von den Driesch is V.O. Vitt (1952) ‘The Horses of the Pazyryk Kurgans’, in Russian, who used GL measurements to group horses into broad size groups – giant, very large, large, medium, middle, fairly small, small, very small and dwarf. Using Vitt’s multiplication factor, the Alde femur falls into the small category (femur GL 35 – 37 cm) again indicating withers height 120- 128 cm, around 12 hands. Johnstone & Albarella (2002) reported finds from a late Iron Age and Roman site in Essex where the mean height of the Iron Age horses was 1260 mm (12.2 hands) and the Roman horses 1376 mm (13.2 hands). Dr Grant sums up: “The picture for horses in England is of general increase in size over time from the early Iron Age onwards in respect of the largest animals, but of much variation at the same time. The conclusion might be drawn … that [the Alde] femur is of a size compatible with animals of the late Iron Age or Roman period”.

FEMUR MEASUREMENTS  (mm)	Alde river	Cambridge Zooarch. lab    #123	NHM   Arabs		Danebury Iron Age (from 5th to 1st cent BC deposits)
Left femur		Donkey	Male	Female	range	mean
GL = ventral surface of condyle to tip of gt. trochanter	355	296.8
L = base of condyle to dorsal surface of femur head	330		436.0	411.0	302.5 - 329	313.8
Median lateral width = post gt. Troch to ant. femoral head	107.5	83.3
Dorso-ventral width of femoral head	41.9	34.4
Medio-lateral width of femoral head	50.0	36.9
Mid-shaft diameter = SD (medio-lateral)	36.2	24.8			33.2 – 37.1	35.7
Medio-lateral width of distal condyle (anterior)	59.8	40.4
Max. width of medial and lateral condyles (posterior)	75.4	62.7

Taking a jump from the reasonable certainties of comparative anatomy, let us envisage a Romano-British scenario. Famously, a bronze head of the Roman Emperor Claudius, unceremoniously ripped from a statuette, was found in the river Alde at Rendham, merely 2 km north of the Straits bridge. It is generally accepted that this relic was loot taken at the sack of Colchester = Camulodunum in the year 61 AD by Boudicca’s Iceni army (Dymond & Northeast, 1995, p. 23). After being carried so far, would such an object be thrown unceremoniously into the river ? Is it possible that an accident occurred at the Rendham crossing. Did a tired horse finally founder at the muddy ford, overturning a chariot and spilling its contents? Hot tempered and impatient to be home, did the Iceni warrior cut free his broken beast, maybe killing it or leaving it to die in the mire? In either case, its bones would slowly work their way downstream with each winter’s flood, until finally being cast up 1946 years later, for the delectation of the river’s mink and the interest of the riparian landowner!

Science, by 14C dating or some other technique, may support or dispose of this hypothesis. For the time being, the femur is stored at Glemham House and available to any interested researcher.

Acknowledgements
I am grateful to three scholarly colleagues, Dr J. Clutton-Brock, Dr A. Grant and Dr P.J. Piper, who kindly provided comparative measurements and comments on equid anatomy. I am solely responsible for the interpretation put on the data.

References
Corbet, G.B., & Harris, S. (eds.) 1991 The handbook of British mammals. 3rd edn. Blackwell

Dymond, D. & Northeast, P. 1995 A history of Suffolk. Revised edn. Phillimore & Co., Chichester.

Grant, A. 1984 The animal husbandry. In B.W. Cunliffe, Danebury, an Iron Age Hillfort in Hampshire Vol. 2: The Finds. Pp. 496-548. London: Council for British Archaeology,

Grant, A. 1991 The animal husbandry. In B.W. Cunliffe, Danebury, an Iron Age Hillfort in Hampshire Vol. 5: The Finds 1978-1988. Pp. 447-487.London: Council for British Archaeology,

Johnstone, C. & Albarella, U. 2002 The Late Iron Age and Romano-British mammal and bird bone assemblage from Elms Farm, Heybridge, Essex (Site code: HYEF93-95). Unpublished Ancient Monuments Laboratory Research Report 45/2002.

Keiswalter, L. 1888, referenced under von den Dreisch, A. & Boessneck, J. 1974 (see text).