10th-11th century Byzantine dromon by Louie da fly - FINISHED - 1:50

[Louie da fly]

This is the very beginnings of a build log. Until I have finished renovating the house, there’s no chance of actually doing any building – no time, and no space available. But in my free moments I’ve been researching and drawing up plans for a Byzantine dromon of the 10^th-11^th century. The name dromon (Greek = “runner”) was originally applied to a class of fast Roman galleys with a single bank of oars developed around the 6th century AD. Over the centuries, as the Roman Empire shifted its emphasis to the East and gained a new capital in Constantinople (modern Istanbul) and evolved into what we now know as Byzantium, the dromon changed as well, until by the 10th century AD it was a very different vessel with two banks of oars, lateen sails and armed with a devastating weapon, pyr thalassion – Greek fire.

 

Greek fire is generally accepted as having been made of naphtha, a naturally occurring substance similar to petroleum. Contemporary descriptions led Prof. John Haldon to work out theoretical design using only technology known at the time, and then put it into practice, with spectacular results (see https://books.google.com.au/books?id=q0hMf5vu7kgC&pg=PA289&lpg=PA289&dq=%22Greek+fire%22+revisited:+recent+and+current+research%22&source=bl&ots=Kwp5Xa3U62&sig=tR81SBsNfAc_uDLyuXDxe9uPWKA&hl=en&sa=X&ei=bjAzVc61C8TNmwWvv4HICw&ved=0CCYQ6AEwAw#v=onepage&q=%22Greek%20fire%22%20revisited%3A%20recent%20and%20current%20research%22&f=false  -

page 292 onward). Greek fire was used effectively in a considerable number of battles, destroying threatening enemy fleets. I was reported to have burnt on the surface of the water (and in the reconstruction that’s just what it does!).

The definitive source on the dromon is the excellent, thorough and painstakingly researched book Age of the Dromon (http://www.brill.com/age-dromon-0) by Professor John Pryor and Elizabeth M. Jeffreys, which draws together all the clues as to the nature of the vessel, from its early development to its apogee in the 10th and 11^th century. The available information is rare, widely scattered and often difficult to interpret. At the time it was written no remains of any Byzantine war-galley had ever been discovered, let alone a dromon, contemporary descriptions were vague and patchy (and in the case of at least one writer of the time, often wrong). Contemporary illustrations are equally unsatisfying and the conclusions and resulting reconstruction reached are a considerable achievement. There have been a lot of theoretical reconstructions of dromons over the years, many of which can be seen if you do a google image search for “dromon”. Some of them are quite ludicrously wrong – often clumsy and far too heavy for a vessel propelled by muscle power. Prof Pryor’s reconstruction is the best and most believable I have seen, and is based firmly on the available evidence. It would also make a fast and effective warship, which most of the others wouldn’t.

 

The picture that emerges is of a long narrow vessel with 100 oars arranged in two banks of 25 on each side. Instead of an outrigger, the sides were flared outwards at the gunwales to allow a good angle of attack for the upper oars. The vessel was fully decked and had two lateen rigged masts and dual side rudders. There was a fortified forecastle, below which was the siphon for Greek Fire. On larger dromons each gunwale was built up amidships and fortified with a wooden ‘castle’.

 

In the centuries between ancient and mediaeval galleys, shipbuilding techniques had changed from frameless shell-construction (in which adjoining planks were fastened together by many small wooden tenons fixed into slots in the edges of each strake), to fully framed construction which was far stronger, and unable to be sunk by ramming. Rather than a ram, the late dromon had a long detachable ‘spur’, intended to break the enemy’s oars, to destroy its motive power and manoeuvrability. Instead of sinking opposing vessels, the dromon’s crew used projectile weapons – bows, slings, ballistas and even flung stones to cause casualties on the enemy’s decks until they could pull alongside, grapple and board. The oarsmen of the upper bank doubled as marines, leaving the lower bank to manoeuvre the ship into position.

 

Tests on Olympias, a full-sized reconstruction of an ancient Greek trireme built in the 1980s, determined that such long, thin, light vessels were very subject to the vagaries of the weather and were of very limited stability. It was discovered that it was impossible to row effectively in any sort of sea – waves one metre high were the upper limit – and a galley faced by bad weather had little choice but to run for shelter. Galleys were also poor sailers and could really only sail effectively in a gentle breeze – a heel of more than 10 degrees would swamp the vessel and it would have been all but impossible to sail into the wind.  A dromon’s storage ability was minimal and on long trips it would have to put into port on almost a daily basis to replenish stores, (particularly water, which oarsmen need to consume in great amounts to stay effective)

 

A galley’s length is determined by the interscalmium – the distance between the rowing benches. Age of the Dromon estimated the interscalmium to be about 1.0 metre (3’3”). Any closer together and the oarsmen foul each other; any further apart and you’re adding dead weight for the oarsmen to pull along for no reason. A galley with 25 benches in each bank would be about 25 metres long, plus extra for the bow and stern. A dromon was incredibly long for its width – the reconstruction in Age of the Dromon is 31.25 metres (95 feet 4 inches) long and its maximum beam amidships only 4.46 metres (13 feet 7 inches).

 

Just as the book was about to be published in 2006 an amazing discovery was made in Istanbul. During excavations for an underground railway in the Yenikapi district of the city’s southern edge, workmen stumbled upon the Harbour of Theodosius, silted up and buried centuries before. Under the mud were the remains of 37 Byzantine ships dating from the 5^th century to the eleventh century AD. Most were merchant vessels, but at least 6 were war galleys – not dromons unfortunately, but galeai – smaller and lighter single banked galleys used for scouting, and from which our word galley comes.

 

The Yenikapi ships confirmed the theoretical dimensions of Age of the Dromon’s  reconstruction – of the two galleys which still had their upper works in reasonable condition, the interscalmium of one varied between 0.90 and 0.97 metres, and the other between 0.874 and 1.048 metres, averaging 0.96.  The length and beam of these galleys was consistent with the theoretical reconstruction.

 

These vessels were incredibly lightly built – the thickness of the planking varied from 20 to 30mm (3/4”-1 ¼”, almost unbelievable in a vessel of that length, and the frame timbers averaged 60mm (2 ¼ ”) square. The ships were prevented from hogging by stringers inside the hull and heavy wales. Presumably dromons were similarly constructed, with extra stiffness provided by the full deck.

 

I cannot adequately express my gratitude for the wonderful help freely given by Professor John Pryor and also by Dr Cemal Pulak of Texas A&M University.

 

Professor Pryor has been incredibly helpful to a lubber like me and I owe to him almost everything I know of dromons, and certainly my decision that I could actually go ahead and make a model of one with a reasonable certainty of getting it right. 

 

He also referred me to his colleague Dr Cemal Pulak, (who took part in the excavation of eight of the Yenikapi ships, including two galleys). Dr Pulak was kind enough to send me a copy of his paper when it was published (it appears in the in the international Journal of Nautical Archaeology 2015  44.1; pages 39-73), as well as a photo of a partial reconstruction of the better preserved of the two galleys he excavated, built at a scale of 1:10.

 

My model will follow Professor Pryor’s reconstruction as closely as possible, with a few modifications based on the Yenikapi finds and on some discoveries of my own when drawing up the model to scale.

 

However, before constructing the full model, I plan to make a midships section at 1:20, with three sets of working oarsmen, as Professor Pryor pointed out that though theoretically the two banks of oars of his reconstructed dromon should not foul each other (Olympias had a lot of trouble with broken oars from fouling between oars of different banks), they have not been tested in the real world. I've attached a PDF of the plans in their current state of development.

 

It should be very interesting.

 

Steven

dromon.dwg Model (1).pdf

[woodrat]

Excellent, Steven. I look forward to following this. I am very jealous that you had access to such eminent specialists in the field.

Dick

[Louie da fly]

Well, actually I just wrote to Professor Pryor and to my surprise he answered. Perhaps that he was at the University of Sydney and that anybody else in Australia was interested in dromons had something to do with it.

 

I've had success with this kind of thing in a previous life as a mediaeval re-enactor, by the way. I've written to the UK, France, even Hungary and got replies - even in one case a copy of a paper written by the person I wrote to. It's been my experience that academics are often quite willing to help someone in the lay public who shares their enthusiasms. I've had my share of 'no replies' as well, but all in all I've been pleasantly surprised by the generosity and helpfulness shown by specialists whom I've asked for assistance. 

 

 

Steven

Edited April 22, 2015 by Louie da fly

[Louie da fly]

I've added a couple more PDFs of the dromon drawings - a transverse section with crew, and views showing just the vessel itself - (no crew, oars, shields etc) showing the shape of the hull more clearly. 

 

The thing is incredibly narrow - at 1:50 it's 62.5 cm ( just under 2 feet) long, but only 9cm (4 inches) wide. The drawings aren't to any particular scale - I had to reduce them to fit on an A3 sheet.

 

As you can see, not much room in the ship once the crew's aboard.

 

I haven't really sorted out the rigging yet - I have more work to put into it, but there's plenty of time for that before I get to the stage where it's needed.

 

Apparently a Greek Fire projector emits a very loud roaring sound when it's in operation. You might note that it's in the shape of a lion, an innovation introduced in the late 11th century by Emperor Alexios I, to further terrify the enemy - as if flames shooting across the waves at you, and the surface of the water burning wasn't enough!

 

Steven

 

PS: Sorry about the pictures being on their side. Just the way they turned out, I'm afraid. 

transverse section crewed.dwg Model (1).pdf

dromon hull only.pdf

Edited April 23, 2015 by Louie da fly

[Matle]

Fascinating subject! Building well-researched models could help sheddibg some light on the construction of these vessels (and a lot cheaper than building full scale replicas, goes without saying). Best of luck!

 

 

Looks horribly unstable though, I have to admit. No wonder they didn't venture far or in bad weather.

[Louie da fly]

Well the Texas A&M University built the bow half of the more complete galley they excavated, but at a scale of 1:10 it took up all the room they had available.

 

I'm afraid my model will have to be POB, as the frames on the original are so frail it would be impossible to do it POF to 1:50 scale (they'd be about 1.2x1.2 mm) and the planking would be 0.4mm thick!

 

But I believe even POB it's going to shed a fair bit of light on how these vessels worked.

 

The whole subject of how they travelled in these things is fascinating. For example, after several hours' rowing, what would they do for toilet breaks? Would there be a rush for the side, unstabilizing the ship? Would they have to have a roster? Who knows . . .

 

Steven 

[Louie da fly]

A couple of days ago I saw  in the archaeological report of Texas A&M University's excavation of ships from the Yenikapi find in Istanbul, something I hadn't noticed before . The drawing of the most intact galley of all, a light galea known as YK4, includes a diagram of the 35th frame of the vessel, which is effectively a midship section.

 

Out of curiosity I printed it off as large as I could, drew up a grid of 1 metre squares around it and copied its shape on my computer to the same scale as the plans for the proposed dromon model, taken from Professor John Pryor's reconstruction.  To compare the two I superimposed the images, and the correspondence was uncanny - at the point they diverge most, the discrepancy between the two shapes is only about 50mm!

 

That a theoretical reconstruction, particularly given the scarcity of information available to base it on, turned out to be so close to the archaeological reality is amazing. Needless to say, I was very impressed, and I take my hat off to Professor Pryor.

 

The only major difference I could see was that the dromon's tholes appeared to be about 170mm higher than those of the ship found in Istanbul. But that was a galea, vessel with just a single bank of oars. A double-banked dromon would be considerably larger and heavier and would sit deeper in the water, so the tholes would be lower. Taking the galea's waterline as being at the bottom of the lowest of the three wales, I re-did the superimposition. They line up perfectly.

 

This has confirmed my confidence in the reconstruction, and that I'll be making something that is very close indeed to the reality of the actual vessel.

 

Steven 

 

 

[Louie da fly]

Here’s a drawing of the proposed midship section for the dromon, to be built at a scale of (probably) 1:20, and intended to sort out whether the upper and lower banks of oars will successfully avoid fouling each other when in action. I haven’t decided how detailed to make it – just a basic mechanical construction to test out the above question, or something fully detailed? After all, the dromon itself is the main goal, but in the meantime I’ve drawn the section in full, in case I choose to do it that way..

 

Though for the full model I’ve followed Professor Pryor’s lead in having the oar benches angled at 18.4 degrees to an athwartships line , I’m still in two minds about it, and I’ve shown the benches in this section as being directly athwartships (i.e. at right angles to the keel).

 

Interestingly, a thole and its “housing” were found in one of the Yenikapi wrecks, but the housing only contained a single hole. Experience with the Olympias reconstruction showed that most of the force expended by the oars at the fulcrum was taken not by the thole pins themselves, but by the leather thong which tied the oar to them. I've now shown the thongs on the drawing, but I haven’t yet added the leather sleeves for the lower oarports, designed to keep water out.

 

As the upper oarsmen doubled as marines during sea-battles, and fought the ship rather than rowing - I've had to figure out what did they do with the oars. The dromon is 4.4 metres wide at its widest point, but the upper oars are over 5 metres long. You can’t pull them inboard – they’d stick out by about a metre either side, making it impossible to come alongside and grapple the enemy (and expose the delicate ends of the oars to damage). Not only that, but it would be murder trying to clamber over them to get from one part of the ship to another – an important consideration during a battle.

 

If you lay the oars fore-and-aft along the deck inboard of the benches, they obstruct access to the benches themselves (the benches are only about a metre apart, so you’d have a stack of oars rolling around getting in the way). The best solution I've been able to come up with – and I’m aware it can’t be proven to be correct – is to have a row of uprights sticking up out of the benches about three oars’ thickness in from the side of the ship. The oars can be racked between these uprights and the side, two or three high, with very little obstruction either for access to the benches or for crew fighting at the side of the ship.

 

Another question I have yet to resolve – was there a catwalk for the oarsmen of the lower bank, or did they simply walk along the tops of the frames? I believe the frames would be close enough together (200-230mm or 10-11 inches) and sturdy enough to be walked on, so a catwalk would be unnecessary.

 

You’ll note the ship has several heavy wales – about 150mm (6 inches) square, to help prevent hogging, and a stringer running along the bottom for the same purpose. These have been found on the Yenikapi galleys, and on a dromon the deck would strengthen the vessel further.

 

There are shields along the sides to protect the upper oarsmen, and it’s mentioned in contemporary accounts that they fought from behind them. The shields are supported by a rail known as a pavesade, probably hung from the pavesade uprights by the enarmes – the leather or rope straps by which the warrior carries his shield in battle. At sea the shield would also have to be tied down further so it wouldn’t flap around with the movement of the ship.

 

Steven

 

PS: The attached drawing has now been amended to show (I hope) the proper relationship of oars, shields etc. 

Edited May 13, 2015 by Louie da fly

[woodrat]

A couple of thoughts. Why would you want to bring the oars inboard during a battle? This would produce lateral instability and a chance of capsize. Bringing the oars onboard would also hinder movement along the deck. Would there be a mechanism to lock the oars outboard to maintain lateral stability and free the rowers to fight. Some illustrations of fighting dromons seem to indicate the vessels going prow to prow and fighting from their castles and siphons. Perhaps boarding occurred over bow or poop.

 

Also would clashing of oars be prevented by increasing the amount of flare of the upper part of the bulwarks slightly as well as having the upper oars longer with different shaped blades because of the lower angle of entry to the water?

 

Cheers

Dick

Edited May 12, 2015 by woodrat

[kees de mol]

Wow very interesting piece of reading you gave us here. I can not help to thing of all those rowers on the galley's becaue I am told they were all slaves. I think it was very hard working on board oft those ships and with high risks if I read of that Greek fire and stuff.

I hope you finish your work on the house fast because I am really looking forward to this build.

 

Regards, Kees

[Louie da fly]

Dick, you make a good point, and certainly renaissance galleys had their oars at "rest" angled upward but supported in their rowlocks. However, they were used in a very different kind of fighting where a galley's function was to sink other vessels with gunfire. They also had only a single bank of oars, which they kept in use during a battle.

 

Mediaeval galleys' crews relied on attrition of the enemy's crew with missile fire, followed by boarding, and after a lot of thought about this, I believe the upper bank oarsmen would have been most likely to put their oars in "storage" to give them the greatest freedom of action. But my proposal is that they are stored running fore and aft right next to the sides of the ship. In this way they shouldn't restrict mobility on board.

 

You're right, though - the approved tactic seems to have been to try to attack an enemy vessel from the quarter, preferably smashing up their oars with the bow-mounted spur, followed by boarding. I can only think the lower oarsmen must have pulled their own oars inboard (they were shorter than the upper ones)  just before impact, but the information available is so thin on the ground that we really know nothing for certain. Even the replacement of the ram with a spur is still not fully accepted by academics, let alone a lot of the other information we have on battle tactics. Much of the contemporary battle advice, such as the treatise of Emperor Leo VI, seems to have been written from the comfort of an armchair by someone who'd never been to sea.

 

The flare of the upper works is in fact what Prof Pryor believes is the only possible solution to the problem of clashing oars. He devotes a whole chapter of Age of the Dromon to this problem, but he acknowledges that it's still only theory till it's tested in the real world.

 

Kees, the oarsmen in galleys of antiquity and the middle ages were free men, and galley slaves don't seem to have been introduced until the renaissance, as a response to a shortage of skilled oarsmen and freemen willing to work at the oars. Rowing a galley was a skilled activity, and good oarsmen were valuable. The most valuable on a dromon were the upper oarsmen, as they doubled as fighting men, and Leo VI recommends that the less brave of the oarsmen be placed on the lower bank.

 

Steven

Edited May 12, 2015 by Louie da fly

[woodrat]

What about the lateral stability provided by oars? This would be lost if all oars are stowed inboard. Would this not increase the risk of capsize? Also, during an action, at least the lower bank of oars would need to be manned in case the dromon had to manoeuvre or in case the upper deck is overrun and the upper bank taken over by the enemy as I believe happened at least once. In that case the lower bank prevented the dromon from being rowed off by the enemy.

Dick

[Louie da fly]

Oh, I wasn't proposing that all the oars would be stored inboard; only the upper bank, and only immediately before and during battle, at which time the masts were also lowered, and the uppr oarsmen became fighters - the lead oarsman on one side becoming (and I love this name) the siphonator, operating the Greek Fire apparatus.

 

As you say, the lower bank would be needed for manoeuvring, and yes, you're right, there's a case on record where the upper deck was overrun and the lower oarsmen weren't, though there are various versions of the story, including one where the lower oarsmen rowed in the opposite direction to the (enemy) oarsmen of the upper bank. This is covered in Age of the Dromon, but Prof Pryor doesn't believe it's physically possible, and that instead the chronicler embroidered the orginal story.

 

Another reason to store the upper oars inboard is that they would otherwise, particularly if fixed in the same way as those in renaissance galleys (see below), get in the way of missiles being fired at the enemy vessel. The lower oars would certainly help provide lateral stability, which would be very important in a vessel as inherently unstable as these.

 

By the way, I'm amending my earlier post to replace the drawing I attached with one which I hope more accurately depicts what I have in mind, including getting the oars and shields in the right relationship to each other, and adding the leather thongs to the thole pins.

 

Steven

 

PS: Kees, I'd also thought that galleys were always rowed by slaves. It took reading The Age of the Galley to change my mind about that.  

[woodrat]

Nice quality drawings, which CAD package do you use? I use Penanink ver 1.1 although I am trying to upgrade to later versions. The lower ports would rquire some modification to  the frames (presumably the toptimbers), to allow excursion of the lower bank of oars. As you are not including this in your model it is probably academic but may be needed in your cross-section mockup. I agree with you that if both banks of oars are manned by one rower per oar, there does not seem to be a need for the oarbenches to be angled and that orientation at right angles to the keel should be OK.

Dick

[Louie da fly]

Thanks everybody for the "likes". They are much appreciated.

 

Dick, I'm using AutoCad LT  2006, which I use in my 'real world' business designing houses. I find it very helpful, but it's only in 2 dimensions (3D AutoCad is prohibitively expensive).

 

I take your point about frame positions, and for the midships section I'll probably come up against this if I decide to do it in full.

 

However, the Yenikapi ships show considerable variation in the spacing of their frames, benches and oarports, and the spacings quoted are only averages. So I might be able to "tweak" the spacings to allow the oarports to go between the frames without playing around too much with the frames themselves.

 

Additionally, there are quite a few frames on each of these ships where the different frame members are next to each other instead of in the same line, so that could be done as well without feeling I'm departing too much from what was actually done. 

 

This has turned out to be quite a challenging build, and I haven't even picked up a piece of wood yet!

 

Steven

[Louie da fly]

Thinking about Dick's question regarding the relationship of the frames to the lower oarports, I've come up with what is probably a workable answer.

 

Working on a spacing of 0.96 metres between benches (as this seems to be pretty much the average found at Yenikapi), that would also mean 0.96 metres between oarports.

 

The average spacing between frames found at Yenikapi seems to be between 200 and 230mm, which is between 1/4 and 1/5 of that distance. If I spaced the frames at 240mm centres, there would be exactly four of them between adjacent lower oarports, and there would be no need to modify the frame spacing to allow for them.

 

With the Yenikapi ships the shipwrights had no need to worry about this issue because those had only a single bank of oars. YK16 had two upper wales with a 55mm strake between them (and I believe YK4 is similar) with the benches mortised into the lower wale, and it appears the thole supports were fixed to the upper wale. As the frames end at the top of the upper wale, there would have been no need to take them into account when spacing the frames.

 

Another point is that this was a  time of transition when ships were being built "shell-first" - that is, the keel was laid and the planks fixed in place before the frames were added. This would have given the shipwright more freedom when deciding on spacing of both oarports and frames.

 

I'm sure when it came to building dromons the shipwrights with several hundred years of experience behind them would have worked out in advance how to place the frames so the oarports would not cause problems. However, as no dromons have ever been found, we can only speculate and use what's been found on the Yenikapi ships as a basis for a 'best guess' design.

 

Steven

 

 

[Louie da fly]

I just came across a very interesting video on Youtube of a reconstruction of Jason's Argo under construction and then in action. Shows the "shell-only" (i.e. frameless) construction of the ship, using coaks between the planks - see https://www.youtube.com/watch?v=-s_0bwC7Hi8

 

Steven

[Louie da fly]

Curioser and curioser - On re-reading the archaeological report of the Yenikapi galleys, I've discovered I'll have to amend my cross-section drawing yet again. It seems the frames in mediaeval (and earlier) Mediterranean ships in general, and in the Yenikapi galleys in particular, followed a

 

 . . pattern of alternating floor timbers and paired half-frames . . . Floor timbers span the bottom of the ship, with their extremities extending just to the turn of the bilge [on each side of the ship]; in contrast, half-fames span the width of the keel and extend up one side of the ship, through the turn of the bilge to, or just beyond, the first wale . . . At each frame station, floor timbers and half-frames are paired with futtocks placed adjacent to, but not fastened to, the floor timber or half-frame, with ends overlapping by the width of one or more planks.   

 

The bit in square brackets is my addition, to clarify that the floor timbers reach right across the bottom of the ship, from side to side.

 

I can't do it at the moment, but I'll amend the drawing to reflect this framing pattern when I get the chance.

 

Steven

[Louie da fly]

Well, here's the amended drawing of the midships section. I've exaggerated the distance between the frames for clarity - with this you can see the alternating bottom timbers and half-frames.

 

I haven't yet sorted out some of the issues, such as how the lower benches are supported at the inboard end - the outboard end is set into a mortise in the wale,  and the inboard end of the upper ones rests on the deck, but I have to work out how to support the ower ones - presumably they'll be supported by the frames.

 

I'm pretty happy with this layout, and I'll probably also see if I can find space to fit such items as water barrels (galley crews needed a LOT of water) within the hull. Prof Pryor's book shows possible places to store stuff, and I have the opportunity to test the theory out in the real world. 

 

The house renovations are nearly complete. I'm hoping that will give me the opportunity to get something started fairly soon - but of course life often gets in the way, so we'll just have to see how it goes.

 

Steven  

Edited May 29, 2015 by Louie da fly

[Louie da fly]

I’ve changed my mind about the midships section – I’m just going to build a “bog basic” section to see if the two banks of oars work together without fouling each other. If so, I can go on with the full dromon with a clear conscience. I may at some future date build the section as well, but I think at the moment doing something that complex is just a distraction from ever starting on the full ship, which is what I intended to do in the first place.

 

I’m having to re-investigate the layout of the ship and may have to re-draw it almost completely, due to the information uncovered by the discoveries at Yenikapi. Though in general these confirm the theoretical reconstruction in Age of the Dromon, there are several important differences between it and the galeai found in Istanbul.

 

Firstly, the reconstruction has the oarports approximately 300mm (1 foot) higher and 300mm forward of the front of the oarbenches. In the only Yenikapi ship (a single-banked galea) with a surviving oarport strake these distances are 400mm and 450-480mm. The discrepancy of height would mean the oars would have to be longer and would reach the water further away from the ship. However, while a dromon has two banks of oars a galea has only one, so the Yenikapi galley had no problems with interference between banks. The nearer the oars are to horizontal to the water surface, the better the mechanical advantage.

 

However, I believe the reconstruction’s oar height is probably closer to the truth, as the shipbuilders would have traded off mechanical advantage against oar-length and height of the upper bank. If a dromon’s lower oars were as long as on a galea, the upper oars would have to be unmanageably long to miss them. And another 100mm (4”) in height would make the dromon that much higher and less stable, and add unnecessary weight.

 

The discrepancy of horizontal offset would mean the oarsmen would have to stretch further to pull on the oars. Without trying this out I don’t know if an extra 150-180mm of reach would be harder on the oarsmen, but as this is something that would have been arrived at after centuries of experiment, I feel it’s likely to be right.

 

Also it appears I've got the midships cross-sectional shape wrong, so I'll have to look at that again and see what I can come up with.

 

 I originally believed the mast would have been permanently set in the mast-step, and pivoted above the deck to lower it for battle. However, on consideration I’ve come to agree with Professor Pryor’s evaluation, that it would have been able to be pulled out with sheers. The masts are fairly light (I worked out a rough estimate based on their length and diameter and the density of the wood likely to have been used), so a reasonably small number of mariners should be able to achieve this.

 

The mediaeval sources describe supports for both the masts and the yards when not in use – they have different names and are almost certainly different pieces of equipment. I have to work out how they would have related to each other on the upper deck. As the yards would be lowered first, I expect the yard supports would be lower, and then the masts lowered onto a somewhat higher set of supports.

 

The sources also describe what appear from the description to be catheads – at the bow, and the anchor hangs from them – though exactly what form they’d take I don’t know. Would they be like later catheads, or would they be very different? Would they have sheaves? I think to do a similar job they’d probably have to be fairly similar in form, but that’s just another of the guesses I have to make in designing this model

.

On the foredeck the ship will need a windlass, a ballista and a siphon for Greek Fire.The sides of the hull near the nozzle had metal plates to protect the timbers. Greek Fire needed favourable conditions – no wind, or a gentle breeze blowing toward the enemy. Otherwise it wouldn’t work, or would be a danger to the operator. It would be nice to have a working Greek Fire unit, but I think that may be a bit foolhardy. I don’t want to set fire to my model by mistake. 

 

The spur was mounted at the bow and was apparently used to smash the enemy’s oars –the ram seems to have vanished several centuries earlier as framed ships were too strong to be sunk by ramming. The design in Age of the Dromon shows a spur coming up diagonally from near the waterline, with a chain or perhaps a metal brace supporting it. But this design seems to be based on a picture of dromon from several centuries before, where the spur appears to be built into the hull like a ram.

 

By the time of my own model the spur was removeable, with an iron fitting holding it in place. If so, ramming would impose considerable torque on the spur, which would tend to pivot around the fitting holding it to the hull, and perhaps seriously damage the ship. I believe the spur should be in line with the force of collision (i.e. horizontal) and would probably fit into a socket in the hull, which is possibly where the “fitting” comes in.

 

I’ve put arched supports for the xylokastron (central castle) – not for any reason other than that the ship is Byzantine and they went in for arches in a big way. As a dromon was the prestigious line-of-battle ship of the time, (the Yenikapi galleys, as Imperial navy ships, were all built of high quality timbers, unlike the merchant ships found in the same site which seemed to be cobbled together from whatever came to hand), I think a bit of ostentation would not be out of place.

 

I’ll need to do some research further down the track about lateen rigging – I really don’t know much about it, and it’s complicated by the fact that there’s pretty much no reliable information about 11^th century Mediterranean lateen rig. Is it the same as later European lateen rigging or is it very different? Some blocks and toggles have been found, and a paper is being written at the moment reconstructing the likely rigging of ships of this time, but it’s not likely to be ready for quite a few years yet.

 

There’s only a fairly narrow access between the oarbenches of the upper deck, and I’ve had a bit of trouble working out where to put the companionway down to the lower deck. I think the most likely place is right by the central mast, but I can’t be certain I’m right. It’s not known for sure how orders were relayed from the captain to the oarsmen, and the lower oarsmen would certainly not be able to hear them directly. On Olympias they used a loudspeaker, but nobody knows what was done at the time.

 

It is recorded that the notes of a flute gave the time to the oarsmen, and perhaps also other commands. Apparently a high-pitched tone carries well among the sounds of a galley under way. However, the sources say this was not done when the ship was going into battle, but they don’t say what method was used. Perhaps messengers sent down to the lower deck to the officers in charge there? Perhaps they used a system of hand signals?

 

Dromons (and all galleys of this type) were poor sailers – their freeboard was so low – a 10 degree heel would put the oarports under the water - and they were so prone to capsize, that they could really only sail before the wind. I think it’s likely that with a following wind the lateen sails of the two masts would have been spread one either side of the centre line, to even up the forces.

 

According to dendrochronology, the galea wreck known as YK4 was perhaps 200 years old when it sank, which is a surprising age for a vessel of this type. It had been repeatedly repaired with different timbers than the ones it had been built from.

 

Hogging was avoided by use of several heavy wales approximately 150x150mm (6”x6”) in cross-section running fore and aft, and all the vessels but YK4 also had stringers running fore and aft fixed to the top of the bottom timbers. A dromon would also have been stiffened by the planking of the upper deck.

One problem that hasn’t yet been solved is how enough air got to the lower oarsmen for them to work efficiently. A fully decked vessel just doesn’t have enough airflow. Perhaps the deck was made of gratings, and perhaps there was some kind of lattice construction between the wales above the lower oarport strake. Nobody knows.

 

The performance of a dromon is unknown. The Olympias, a reconstruction of an Ancient Greek trireme, was capable of 7 knots and it’s thought if another one was built today using the lessons learnt from the first it could go even faster. Olympias is capable, at full speed, of turning in twice its own length, an impressive feat. I doubt a dromon could achieve this performance, but it was the racehorse of its day. It was superseded by the more efficient Western European galea (not the same as the Yenikapi galea, from which it got its name), which rowed two banks of oars from a single set of benches. It was lighter and faster than the dromon, and edged it out in little more than a century.

 

To my surprise I discovered that side rudders (steering oars) are more efficient than a stern rudder is. There’s a fascinating paper by Lawrence Mott on the development of the rudder from 100 to 1600 AD at http://nautarch.tamu.edu/pdf-files/Mott-MA1991.pdf which goes over this in great detail. Well worth a read. The only reason I can think of that the stern rudder would have replaced the side rudder is the increasing size of ships which would have needed a steering oar so big it was impossible to handle.

 

Most Mediterranean ships of the time had two steering oars, which could be raised and lowered independently of each other, so the ship could proceed with a single rudder if needed (there were times when this was a better option). They needed to be held down or they’d float upwards out of their proper position. They could also be raised right out of the water when (as was usual with galleys) the ship was beached overnight, aft-first.

 

As I mentioned before I worked out a rack system to hold the upper oars for when the ship went into battle. Each upper oar is a little over 5 metres long and the distance between the oarbenches is a little under a metre. So at any one point there are 5 oars in the rack. It seems to me that placing the oars in the rack would have to be done very systematically, otherwise chaos would ensue, with oars getting mixed up and impossible to extricate later. I think they’d have to be laid “spoon-fashion”, probably starting with the bow-most oar and working down the side of the ship. Picking them up would be done in reverse order.

 

Olympias showed it was vital for oarsmen to take their seats in the proper order, especially for the lower decks, and the same should apply for a dromon. Pulling the lower oars inboard (say when coming alongside an enemy vessel to board) should be reasonably ok – I worked out if you pull the oar inboard on a diagonal there should be room – but there might be problems with the foremost and aftermost oars. Getting the oars below decks down the companionway might also be a problem

.

There appear to have been two major forms of tactics – starting with a crescent-shaped line abreast facing the enemy, the idea was either to break the centre of the enemy line or to go around the end of their line and attack from the stern. Either way, the idea was to break up their formation and the battle apparently then degenerated into a melee of individual ships. However, as with everything else, the information is so scarce that nobody really knows. It was advised to fight near land – but not your own territory, as the proximity of friendly land made desertion more likely.

 

I’ve worked out that the deck beams need some kind of support part way along their length, as timbers that could cover 4 metres in a single span would have to be far too heavy for a ship of this size. However, a central beam, or perhaps two beams equally spaced, running fore and aft under the deck beams would do the job. Vertical struts at intervals, supported by the ship’s frames or keel would hold up the fore and aft beams. It would be preferable to have two beams, to avoid a line of vertical struts running down the centre of the ship obstructing movement. But if there were two they’d have to be positioned so the struts didn’t get in the way of the oarsmen when they were rowing.

 

Well, that’s the current state of play. I’m looking forward to getting started, but I think it’s going to be a while yet.

 

Steven    

Edited July 8, 2015 by Louie da fly

[Louie da fly]

Here are the two earliest representations I can find of mediaeval lateen rigging that show any detail at all. The earlier one is 9th century AD Byzantine and is fairly simple, but it does show the shrouds and the fore tackles for the yard, as well as what appears to be a halyard with a block. The rudder's also very interesting - it appears to be attached to a platform.

 

The other is from 1339 and is Italian, and has a LOT of detail, including not only the shrouds but their method of fastening -not deadeyes but blocks [edit - this is so the shrouds on the lee side of the ship could be slackened off when tacking, so they didn't keep the sail from filling with wind to its fullest extent - see attached picture] - a rope ladder to the top, fore tackles, bolt ropes and even the shape of the tiller and the tackles that hold the rudder in position, one fore and one aft.

 

It also shows the shape of the top and apparently how the shrouds are fixed to the mast (wrapped around) - I have no idea if this was common practice in the Middle Ages, but it's certainly different from later practice [edit - I've since found at least one other representation of the tops of the shrouds wrapped around the mast just under the top]. The other details of the ship of course have very little to do with a dromon - totally different type of ship, and several centuries too late. But one question - what are those ropes halfway along the yard that seem to lead upwards to heaven?  

Edited July 8, 2015 by Louie da fly

[Louie da fly]

Some more mediaeval pictures of ships with lateen sails.

 

The first is Catalan, thought to be from the mid-late 13th century (maybe about 1290) and shows quite a bit of detail - shrouds, halyards, braces and perhaps a fore-tackle, plus what may be a parrel truck fixing the yard to the mast - or perhaps just a loop of rope (sorry about the watermark - this is the biggest representation I could find on the net).

 

The next is mid-14th century Italian and shows most of the ropes and blocks. The fore-tackle is particularly clear.

 

The next is Byzantine, from about 1200 AD and again shows shrouds, tackles and braces. The detail's not as good, as the picture's so small.

 

The fourth picture is 14th century Spanish. It shows a single brace and fore-tackle, and halyards, but no shrouds. The yard seems to be held to the mast with a double loop of rope.

 

But a picture's emerging which should allow me to produce a rig which I'm confident is close to what was in use at the time. There seem to have been variations, as one would expect - the fixing of the shrouds to the mast, the suspension of the yard, either by a loop of rope, or two loops. I've also come across pictures of lateeners with parrels, but I don't seem to have taken copies (damn!). Though I'd expect larger ships to use parrels, I haven't got any hard evidence for them in the right period, so I probably won't have them on the model.

 

There seems to be no sign of lifts for the yard at this time, and it's perhaps debatable whether they could have been used, as the yard seems almost always to have been hoisted all the way to the top of the mast.

 

Also, a picture by a Byzantine artist of a galley which appears to me to have a horizontal spur at the bow, just as I proposed in my last post. Unfortunately I don't know the source, but it looks to be from some time in the 12th century AD. The artist seems to have got the direction of the oarsmen wrong - they should be facing the stern, which is where the steersman is.

 

And finally, a Spanish picture from between 1270 and 1284 which shows a galley in the foreground with what appears to be a lowered mast (almost horizontal, in front of the soldiers). The ships in the background still have their masts vertical.

 

The "wings" of the ship are at the stern, so the mast is leaning aftwards. If this really is a mast, I may have been wrong in thinking the masts were pulled out of the deck with sheers - I may have to go back to the idea of a pivoting mast. Of course, it's possible that what we see here is not a lowered mast but a broken one - the lower end does look like it may have been broken off.

 

Aha! I've found the other picture of the tops of shrouds wrapped around the mast - it's early 15th century, a galley from Michael of Rhodes' treatise on shipbuilding. See the final picture. It's also got some very good details of rigging, though I have to say it's rather late - up to 400 years after the time I'm interested in, and a lot of evolution would have happened in ship design in the meantime.

 

Steven

Edited July 8, 2015 by Louie da fly

[woodrat]

These references may be of use:

 

Pryor, J Mariner's Mirror 219: 70.3 fig.13

Also reproduced in Lillian Ray Martin

 

also Landstrom, Bjorn 1961: the Ship: an illustrated history figs. 215-18

 

Dick

[Louie da fly]

Thanks, Dick. I'll try following up on these. I've got Landstrom's book and I've been looking at his illustrations and text. I'll see what I can do about the others.

 

I think in general I've now got a reasonable understanding of how this rig worked and what it was made up of, which I certainly didn't have before. But I'll probably learn more as I start building. 

 

Steven

[Louie da fly]

Here’s the latest on the design process. I’ve changed the cross section in line with the discoveries at Yenikapi, and it’s made quite a difference in many areas. The underwater hull has a considerably flatter bottom than the previous reconstruction, and this means the underwater shape is different.

 

Effectively, the ship sits lower in the water, and this affects the height of the lower bank of oars – from what I’ve been able to find out, a freeboard of about 1 metre (3 feet) was usual between the water and the lower bank of oarports. So the oarsmen are going to be sitting higher in the ship. And as you need about a metre between the heads of the lower oarsmen and the upper benches, the upper guys move up too. And then the floor structure of the midships castle, which needs to clear the heads of the upper oarsmen, has to be higher. Everything affects everything else . . .

 

I’ve done about half the body plan -  from the bow to midships (attached) showing the assumed position of the oarsmen. But of course, how low in the water the ship actually sat is an imponderable as well, depending on the weight of the ship itself, plus that of the crew and cargo and equipment etc etc. I can only guess and hope I’m about right. What appears to be the upper part of the hull is very lightweight - just a rail (pavesade) to carry the shields of the upper oarsmen. But the midships castle would weigh a fair bit, so how far down the ship would sit is rather uncertain.

 

I’ve also changed a few details. The ship is “leaner” in sheer view, and I’ve altered the shape of the  “tail” and that of the side rudders to be more in line with illustrations contemporary with a dromon of the period of the Macedonian Emperors of the 10^th and 11^th centuries AD.

 

The spur at the bow is fixed higher and is horizontal rather than angled upward, to avoid the tendency to pivot around its point of attachment on impact, with the danger of damage to the bow of the ship.

 

There’s more sheer, and the bow is wider and shallower to ride over the sea rather than cut through it, in line with the Yenikapi finds, as shown on the photo below of a 1:10 scale reconstruction done by TAMU . However, this is of a single-banked galea - no dromons were found at Yenikapi, so I've had to adapt this to a two-banked design and hope it's right.

 

Generally, however, it’s still very close to Prof Pryor’s original concept.

 

I’m still working at it, but it’s all rather slow. 

[Louie da fly]

Well, I’m finally back onto the dromon after a lot of waffling about trying to work out how best to build it.

 

The true shape of a dromon’s  hull is unknown and my build will be based on the theoretical reconstruction in Professor  John Pryor’s book Age of the Dromon, plus very valuable insights from  the wrecked galleys found at Yenikapi in Istanbul. These vessels are not dromai but the smaller lighter galeai, and I’ve extrapolated from these to get the shape of the larger vessel.  It will evolve very much by trial and error.

 

The Yenikapi galleys were incredibly lightly built – necessary when ships have to be powered by human muscle. Though they were about 30 metres (90 feet) long, their keels are made of timbers only 150mmx150mm (6”x6”), and the planks are only 20-30mm (3/4”-1.2”) thick.

 

I’ve finally decided to build it plank on frame, but as the planks are going to be 0.5mm (1/50 inch) thick and the frames 1mm (1/25 inch) square, I’ll build it around a plug, which I’ll then remove and insert the frames within the hull. Otherwise I’m likely to smash the planking when I pick up the ship to work on it.

 

Dromons came at a time when ships were in transition from being built planking first to frame-first, so I’m sort of following in their footsteps.

 

I’ve been inspired by Druxey building his Greenwich barge around a plug, which has given me lots of ideas for my own build ( though I don’t think I’ll ever approach his quality of work). I’d already started making a plug out of a single piece of pine, but realised I had no idea how to get the two sides of the hull symmetrical.

 

 

 

Taking a leaf out of Druxey’s book I’m going to make a new plug out of narrow pieces laminated together horizontally, mirrored each side of the keel, which should pretty much solve that problem.

 

 Unfortunately though the keels of the Yenikapi galleys have been found the stemposts and sternposts weren’t recovered, so I’ve had to base mine on contemporary pictures.  I’ve made these, along with the keel, which is 3mmx3mm (1/8” square) in the model.

 

 

 

 

A keel this thin is very given to warping and I found when I’d made mine it was a bit bowed sideways. I tried putting it in boiling water and then between two sheets of glass, but though that helped reduce the bow, it also made the keel hog, so I’ll have to try another tack – steaming and clamping.

 

Once that’s done I’ll be putting the plug together and shaping it.

Edited February 13, 2016 by Louie da fly

[druxey]

Steven: see my comments in my Greenwich Hospital barge log. With carvel planking, edge glued planks will not hold together as they do in a clinker build! There is insufficient gluing area for this. You'll also need to devise some method to stabilize the thin keel to keep it straight until the planking is on.

[woodrat]

Just a suggestion but putting the frames external to the plug as here in my longboat  http://modelshipworld.com/index.php/topic/4496-usf-essex-by-woodrat-scale-1-64-fully-framed-from-takakjian-plans/?p=142797 allows you to edge glue the planking and fix the planks on the frames. The keelson ( if dromons had them) can be morticed to hold the frames and the keel fixed over the frames.

Dick

[druxey]

Dick: I had no trouble planking with the frames recessed in the plug when the plug was well waxed.

[Louie da fly]

Thanks everybody for all the 'likes', and especially to Dick and Druxey for the kind advice on the best way to plank the ship and put in the frames. I'm still feeling my way with all this - and recessing the frames might be the way to go. Druxey, I take on board your comment about  stabilizing the keel - it's been one of my major concerns. Dick, yes the Yenikapi galleys do have keelsons. It's going to be important because of the unusual method of framing in Byzantine ships - alternating frames starting from the keel with frames that crossed it. They're very close together because they're so light, so I'll have my work cut out for me. It's going to be like building a 30 metre longboat. 

 

I'm trying to approximate as closely as possible the timbers used on the original Yenikapi galleys, to get a (very) rough idea of how much one of these things would have weighed. The keels and frames seem mostly to be made of oak or of European plane and the planking is of black pine. There's ample oak and plane trees here in Ballarat and I follow the council workers around when they lop the street trees. So the keel, stempost and sternpost are of plane which I picked up off the side of the road, and I'll make the frames of it too. I can't get black pine, so I'll probably fudge it with radiata - cheap and easily available here.

 

Steven

Edited February 9, 2016 by Louie da fly