Could the Persian Empire logistically support an army several million strong?

[Erpingham]

It's looking like food was taken to horses, not vice versa. Would the same apply to pack animals? I suppose it would. How would a regular army like Rome's feed its animals on the march?

Yes, you'd need to deal with pack animals similarly, for reasons already given.  You need to be able to be able to round up your animals and get them loaded in reasonable order at the beginning of the day. 

I've looked at some of the figures in the Crusades logistics book and compared with what is said about hay yields on the internet.  There is a huge difference in yields over time and space but essentially modern hay yields are much higher than historical ones, so we should try to go that way.  There is also some confusion on whether yields are per year or per cut, and how many cuts per year.  Round here they cut twice.  From my quick viewpoint 1.0-1.5 tonnes per acre for a cut would probably bracket it.  1 tonne would feed 100 horses for a day, proportional to the amount of dry fodder put in.  I can't remember how many horses, mules and camels we have.

[Jim Webster]

It's looking like food was taken to horses, not vice versa. Would the same apply to pack animals? I suppose it would. How would a regular army like Rome's feed its animals on the march?

Yes, you'd need to deal with pack animals similarly, for reasons already given.  You need to be able to be able to round up your animals and get them loaded in reasonable order at the beginning of the day. 

I've looked at some of the figures in the Crusades logistics book and compared with what is said about hay yields on the internet.  There is a huge difference in yields over time and space but essentially modern hay yields are much higher than historical ones, so we should try to go that way.  There is also some confusion on whether yields are per year or per cut, and how many cuts per year.  Round here they cut twice.  From my quick viewpoint 1.0-1.5 tonnes per acre for a cut would probably bracket it.  1 tonne would feed 100 horses for a day, proportional to the amount of dry fodder put in.  I can't remember how many horses, mules and camels we have.

If somebody can tell us the number of animals, nutrition requirements is just a matter of calculation :-)

[Erpingham]

If somebody can tell us the number of animals, nutrition requirements is just a matter of calculation :-)

Very true.  I thought presenting this in such a generic way would be more useful to Justin.  Whether Justin can create anything more plausible by using them, I don't know.

[Justin Swanton]

It's looking like food was taken to horses, not vice versa. Would the same apply to pack animals? I suppose it would. How would a regular army like Rome's feed its animals on the march?

Yes, you'd need to deal with pack animals similarly, for reasons already given.  You need to be able to be able to round up your animals and get them loaded in reasonable order at the beginning of the day. 

I've looked at some of the figures in the Crusades logistics book and compared with what is said about hay yields on the internet.  There is a huge difference in yields over time and space but essentially modern hay yields are much higher than historical ones, so we should try to go that way.  There is also some confusion on whether yields are per year or per cut, and how many cuts per year.  Round here they cut twice.  From my quick viewpoint 1.0-1.5 tonnes per acre for a cut would probably bracket it.  1 tonne would feed 100 horses for a day, proportional to the amount of dry fodder put in.  I can't remember how many horses, mules and camels we have.

If somebody can tell us the number of animals, nutrition requirements is just a matter of calculation :-)

Someone want to do the maths? I'm feeling lazy.

[Erpingham]

To take us further into the fog of agricultural production, timed tests of Indian sickle harvesting on line show about 100 square metres per hour.  So an acre could be harvested in 40 hrs, or 40 people could harvest the acre in an hour.  The killers (regardless of number of horses) would be distance needing to be travelled to the pasture and the productivity of the pasture when the harvest team arrived.  Given all the other calls on the horse-support team, relying on fodder carried with the unit and available in its horse lines makes sense.

[Justin Swanton]

OK, let me do the maths.

Take 1 ton per acre which feeds 100 horses. There are 2,47 acres to a hectare so assume one hectare feeds 250 horses.

Now let's assume that the army, at least whilst marching inland, feeds its animals entirely from fodder collected from the land.

Each man needs 1kg of grain a day. Each mule can, at a cautious estimate, carry 250 pounds of grain =  113kg, call it 100 kg.

The army marches inland for 5 days tops. So each mule must carry 5kg per man, which works out to 1 mule per 20 men.

The army numbers 3 400 000 men. That's 170 000 mules. Tack on another 20 000 horses or so for the cavalry. 190 000 animals. Call it 200 000 animals

Assume a mule eats as much as a horse. That means you will need 800 hectares of fodder each day.

The army pitches camp. Working on 1000 men per hectare it occupies an area of 3400 hectares.

40 people can harvest an acre in an hour so 100 people can harvest a hectare in an hour. 80 000 people can harvest 800 hectares in an hour.

Following these figures, the time spent preparing the camp will be enough to gather about four times the fodder requirements for the animals without the gatherers having to stray beyond the camp limits.

Any chinks in the argumentation?

[Jim Webster]

One problem is that you're assuming  constant hayfields

Firstly I think you have to exclude the camp from the calculation, because that will be full of people doing stuff and once the first units have arrived they're going to trample and soil things. There could be a bit from the site but it'll probably end up fenced for for the private horse herds of guard units and the the great man's chariots etc

Secondly unless land is actively fenced off to ensure that the grass is protected, it doesn't end up the length you are talking about. Local livestock will have taken it long before.

I found a picture entitled, Europe, Grece, Plain of Thessaly, Valley of Penee, herd of goats -

Given the colour it's in the fertile time of the year. Even if there were no goats, you're not harvesting grass from endless plains.
You'll probably  need an order of magnitude more area to get the amount of grass you need and that will be in the good times. When it's earlier in the season you'll actually take longer because the grass is shorter so it's harder to cut. It's also lower dry matter so you'll need more weight.

with regard to the men doing it, every army I've ever read about, each animal will be fed as an individual or as a team. So 200,000 animals will probably have over 100,000 men out there getting the fodder for the animal/s they are personally responsible for.

[Erpingham]

Using the camp area to forage from is quite neat Justin - novel.  Presumably, there would always be a camp construction unit a day ahead of the main army who would stack the fodder for the next day?  You'd need different arrangements for the once a week depots, if still having them.

The biggest issue is really Jim's - can you find 3,400 hectares of good quality hay field every 24 hours?  At one point, you're crossing a mountain, another carving through a forest and a third you are in a concentration area for several days. Was the rest lush, unutilised grassland? 

The other issue I think I spot is the number of horses.  I thought we started with either 80,000 cavalry? According to Iranica online, there were also 20,000 charioteers (assuming two man light chariots, that 20,000 horses) and some camel cavalry as well.

When we've done this bit, maybe we could talk naval logistics?  Is the fleet using the same depots?  Assuming the sealift conveyor can carry their own crews supplies (which seems reasonable), Iranica reckons there were about 250,000 rowers and marines.  Rowers probably had a higher ration rate than the common herd, because of their job.

Finally, a bit of trivia.  You could fit Yasgur's Farm 14 times into Justin's camp

[Justin Swanton]

One problem is that you're assuming  constant hayfields

Firstly I think you have to exclude the camp from the calculation, because that will be full of people doing stuff and once the first units have arrived they're going to trample and soil things. There could be a bit from the site but it'll probably end up fenced for for the private horse herds of guard units and the the great man's chariots etc

Secondly unless land is actively fenced off to ensure that the grass is protected, it doesn't end up the length you are talking about. Local livestock will have taken it long before.

I found a picture entitled, Europe, Grece, Plain of Thessaly, Valley of Penee, herd of goats -

Given the colour it's in the fertile time of the year. Even if there were no goats, you're not harvesting grass from endless plains.
You'll probably  need an order of magnitude more area to get the amount of grass you need and that will be in the good times. When it's earlier in the season you'll actually take longer because the grass is shorter so it's harder to cut. It's also lower dry matter so you'll need more weight.

with regard to the men doing it, every army I've ever read about, each animal will be fed as an individual or as a team. So 200,000 animals will probably have over 100,000 men out there getting the fodder for the animal/s they are personally responsible for.

A properly cultivated field can give you 10 tons per acre, so 1 ton per acre for unimproved land seems reasonable to me, and the army will be camping on flat ground (most amount of soil, least amount of rock - optimal for grass) pretty much all the time.

I think it natural that the army high command will order foragers to reap the grass from the campsite before the arrival of the main body of men. They don't want to waste time having cast around beyond the camping area for the fodder.

One must also factor in time. From April to September there are at least 15 hours of visibility per day. At a steady 3km/h, the army can cover 20km in 7 hours. Add a couple of hours for breaks, delays and general faff and the army is on the road for 9 hours. That leaves 6 hours of daylight for non-marching activities.

One thing which does queer the pitch somewhat is the length of the column. Presuming the army marches 300 men or 600 yards wide, the column will be somewhere in the region of 23km long if each 'rank' is 2 yards deep. This means that if the head of the column sets out at 6 am, say, it will reach the next campsite at 3 pm without the tail end of the column having yet left camp. Presuming the army marches only during daylight, it has 15 hours of marching time. The entire column must be able to pass a given point in half that time - 7,5 hours, for all the army to reach the next campsite before dark. Which suggests either that the men don't stop marching for any reason during the entire day (eating and drinking what they carry with them) or that the column is substantially wider, say 450 men on a frontage of 900 yards. Final option: the army doesn't cover more than about 12 km in a day. It is a poser.

[Erpingham]

A properly cultivated field can give you 10 tons per acre, so 1 ton per acre for unimproved land seems reasonable to me, and the army will be camping on flat ground pretty much all the time.

Just a reminder.  You have to remember yields have risen a bit since Xerxes day and that Northern Ireland isn't Northern Greece.  As suggested, I'd look at a range of historical yields, rather than modern ones.

[Jim Webster]

One problem is that you're assuming  constant hayfields

Firstly I think you have to exclude the camp from the calculation, because that will be full of people doing stuff and once the first units have arrived they're going to trample and soil things. There could be a bit from the site but it'll probably end up fenced for for the private horse herds of guard units and the the great man's chariots etc

Secondly unless land is actively fenced off to ensure that the grass is protected, it doesn't end up the length you are talking about. Local livestock will have taken it long before.

I found a picture entitled, Europe, Grece, Plain of Thessaly, Valley of Penee, herd of goats -

Given the colour it's in the fertile time of the year. Even if there were no goats, you're not harvesting grass from endless plains.
You'll probably  need an order of magnitude more area to get the amount of grass you need and that will be in the good times. When it's earlier in the season you'll actually take longer because the grass is shorter so it's harder to cut. It's also lower dry matter so you'll need more weight.

with regard to the men doing it, every army I've ever read about, each animal will be fed as an individual or as a team. So 200,000 animals will probably have over 100,000 men out there getting the fodder for the animal/s they are personally responsible for.

A properly cultivated field can give you 10 tons per acre, so 1 ton per acre for unimproved land seems reasonable to me, and the army will be camping on flat ground (most amount of soil, least amount of rock - optimal for grass) pretty much all the time.

I think it natural that the army high command will order foragers to reap the grass from the campsite before the arrival of the main body of men. They don't want to waste time having cast around beyond the camping area for the fodder.

One must also factor in time. From April to September there are at least 15 hours of visibility per day. At a steady 3km/h, the army can cover 20km in 7 hours. Add a couple of hours for breaks, delays and general faff and the army is on the road for 9 hours. That leaves 6 hours of daylight for non-marching activities.

One thing which does queer the pitch somewhat is the length of the column. Presuming the army marches 300 men or 600 yards wide, the column will be somewhere in the region of 23km long if each 'rank' is 2 yards deep. This means that if the head of the column sets out at 6 am, say, it will reach the next campsite at 3 pm without the tail end of the column having yet left camp. Presuming the army marches only during daylight, it has 15 hours of marching time. The entire column must be able to pass a given point in half that time - 7,5 hours, for all the army to reach the next campsite before dark. Which suggests either that the men don't stop marching for any reason during the entire day (eating and drinking what they carry with them) or that the column is substantially wider, say 450 men on a frontage of 900 yards. Final option: the army doesn't cover more than about 12 km in a day. It is a poser.

I) you're not talking about 'unimproved grassland' Much of Greece doesn't even aspire to the title grassland.

2) The army 'high command' has nothing to do with the foragers. The foragers are the private servants of the individual soldiers. They're not an army unit for the high command to order about.

3) marching. Rule of thumb here. Oxen can work for 8 hours, graze for 8 hours, ruminate for 8 hours. Camels ruminate as well as oxen. Horses don't but they cannot cope with working and an all grass diet. So forget having your baggage march for 20km a day

4) the maths are beginning to catch up with you. You're now beginning to see why very few believe that Xerxes lead an army that large. For large parts of the journey, for example Gallipoli, there is nowhere you can march 300 men wide. Basically a lot of men are going to be camping where other men camped the night before.
Rather than thinking of an army leaving camp, marching and entering camp, with the sort of numbers Herodotus was talking, it makes more sense to consider the army as a camp which is moving because only the lead units will ever camp of ground that hasn't been camped on

[Jim Webster]

A properly cultivated field can give you 10 tons per acre, so 1 ton per acre for unimproved land seems reasonable to me, and the army will be camping on flat ground pretty much all the time.

Just a reminder.  You have to remember yields have risen a bit since Xerxes day and that Northern Ireland isn't Northern Greece.  As suggested, I'd look at a range of historical yields, rather than modern ones.

Northern Ireland is one of the best grass growing areas in the world. Parts of New Zealand might be better 

[Justin Swanton]

I) you're not talking about 'unimproved grassland' Much of Greece doesn't even aspire to the title grassland.

Here's something interesting, a paper on grazing land in Greece. Page 4 describes the different yields between poor brushland and improved grassland. The poorest brushland (at 100% cover) yields 762kg of fodder per hectare whilst good quality grassland yields 4500kg/ha - which comes out at a ton per acre for average land. This means that poor land in Greece will feed 70 horses per hectare or 30 horses per acre (or somewhat less since brush is not as nutritious as grass). At the worst campsite harvest everything in the camp and a little beyond and the animals will be fed.

2) The army 'high command' has nothing to do with the foragers. The foragers are the private servants of the individual soldiers. They're not an army unit for the high command to order about.

Then Xerxes orders that a foraging corps be organised that collects the grass/brush at each site. it must be done so it is done.

3) marching. Rule of thumb here. Oxen can work for 8 hours, graze for 8 hours, ruminate for 8 hours. Camels ruminate as well as oxen. Horses don't but they cannot cope with working and an all grass diet. So forget having your baggage march for 20km a day

12km a day?

4) the maths are beginning to catch up with you. You're now beginning to see why very few believe that Xerxes lead an army that large. For large parts of the journey, for example Gallipoli, there is nowhere you can march 300 men wide. Basically a lot of men are going to be camping where other men camped the night before.

12 km a day.

I dispute Gallipoli. I mapped out a route that allows for a wide column to move up the peninsula. Shall I post photos?

[Jim Webster]

I) you're not talking about 'unimproved grassland' Much of Greece doesn't even aspire to the title grassland.

Here's something interesting, a paper on grazing land in Greece. Page 4 describes the different yields between poor brushland and improved grassland. The poorest brushland (at 100% cover) yields 762kg of fodder per hectare whilst good quality grassland yields 4500kg/ha. This means that poor land in Greece will feed 70 horses per hectare or 30 horses per acre (or somewhat less since brush is not as nutritious as grass). At the worst campsite harvest everything in the camp and a little beyond and the animals will be fed.

you missed this bit about the uses of the brushland
" Goat raising, and to some extent sheep raising also, depends greatly on browse and grass produced on evergreen brushlands. This is
especially true for the period from late autumn to late spring, when the upland ranges are not ready to be grazed."

The local livestock have been wintered on this land already, there's damn all there and what there was was just about suitable for Goats

[Patrick Waterson]

Getting back to the number of animals, we have for the Achaemenid mounted component in Herodotus VII.86-87:

>80,000 horsemen, consisting of Medes, Cissians, Bactrians, Caspeirians and Paricanians, plus some Indians. 
>Libyan and Indian chariots, number unspecified, plus Arabian camelry, number unspecified.

In VII.184 Herodotus reckons the chariotry and camelry together as 20,000 men.  The chariots would be multi-man, with 2-3 per Libyan chariot and perhaps 3-4 per Indian chariot, while the camelry could be single- or double-mounted.  We are probably looking at something like one man per mount, so a grand total of 100,000 combat animals with 100,000 mounted troops.

Baggage animal numbers depend heavily on forage considerations.  If forage can be cut and stocked locally, then we are looking at only enough baggage animals to carry rations for the men (and presumably camp followers, assuming these are intended to survive).  Leaving aside meat on the hoof, and taking Herodotus' minimum of one choenix (about 2 lbs or slightly under 1 kg), the 3.6 milion souls in Xerxes' army would have a daily requirement of 7.2 million pounds of grain.

If we use Anthony's latest finding for General Crook's 400 lbs per mule, and for simplicity let the camels carry a similar load, we get 18,000 animals per day's capacity of food.  Giving the baggage train one week's supply capacity requires 126,000 animals, which we can take as the minimum number.

Any addition of animals over and above this represents 1) meat on the hoof, which where possible will be grazed, and 2) fodder carriers.  The latter will be required only where fodder is inadequate, and given that Xerxes' march as far as Thermpoylae was principally through territory where he had ordered the population to stockpile and provide food for his army, which they did (unless we are going to start disputing Herodotus' account of this point), there is every reason to suppose they were collecting fodder, too.

In essence, this would mean that the Persians would not need a massive addition of fodder-carrying animals to their baggage train, and thus would substantially avoid the vicious cycle whereby as soon as one starts adding fodder-carriers one needs to carry fodder for the fodder-carriers.

So we are looking at a baggage train in the order of 126,000 animals, not far off my earlier estimate of 135,000.   To give 3.6 million men a meal at 50 men to the ox, assume 72,000 cattle are tagging along as well at any one time, for a total fodder requirement of:

>100,000 combat mounts
>130,000 (rough average of 126,000 and 135,000 for convenience) baggage animals
> 72,000 cattle for consumption

This is about 300,000 animals overall.  If the obedient locals through whose lands the army is passing are gathering much of the fodder, as seems reasonable in view of their having fed the Achaemenid army while it passed through, then the army itself is spared much of the drudgery of collection and has only the task of distribution.