March 1814. The French cavalry were broken. Streaming back to the bridge at Arcis sur Aube only the presence of the Emperor and a few battalions of the Old Guard had them starting to rally. An enemy shell landed a few feet from the shaken troops causing a general withdrawal from the area of its smoking fuse and threatening a panic stricken retreat, some men even pushing their way through the ranks to avoid the coming explosion.

Napoleon acted quickly to quell the incipient retreat and forced his horse over the shell. It exploded and disembowelled the horse. The Emperor disappeared in a cloud of smoke, but then got up, unharmed and mounted another horse'. His example steadied the wavering cavalry, but more to the point of this article it demonstrated the relative weakness of a shell burst.

A similar weakness was shown up by Captain Hay's experience at Waterloo where a shell burst under his horse, but only hurt his horses rear legs. Also at Waterloo a shell landed near Wellington and blew several grenadiers in a nearby square to pieces. Not so innocuous! A last comment on innocuous fire comes from Wagram where dozens of Austrian shells landed in the Saxon bivouacs on Lobau, but the Radeloff Grenadiers only suffered 20 casualties.

Indeed Mercer at Waterloo remarked on the shells "falling amongst us with very long fuses, kept burning and hissing a long time before they burst, and were a considerable annoyance to man and horse." One shell exploded under two horses and they both dropped dead. Overall the troop lost 150 horses out of 226, though it is not clear which types of projectile caused this loss.

Sergeant William Lawrence in his autobiography in 1856 wrote that he saw a shell cut his sergeant major in half, take off the head of a grenadier, and then explode and blow himself off his feet.

During the horse and musket period being examined in this series shells were round missiles with a thin iron skin (see sketch 1). Because of the thin skin it was not used in guns whose relatively high initial velocity would have burst the projectile in the barrel, so shell was only fired from howitzers and mortars. It was discovered after 1815 that a shell could be fired from a gun using a reduced charge, but this was not the practice during the Revolutionary and Napoleonic Wars.

A howitzer is a cross between a gun and a mortar, and has a much shorter barrel than a gun. It was normal for a 6 to 8 piece battery to include 1 or 2 howitzers. Unusually both the British and Prussians on occasion fielded all howitzer batteries.

Effective Ranges

The effective range of this type of projectile was between 700 and 1200 yards because of the length of the fuse and the uncertainty of its exploding. The minimum range was because some of the fuse, which was ignited inside the barrel by the propellant charge, had to be left outside the shell. (See sketch 2). However, de Toussard 13 writes how effective 6.52 inch howitzers were in the use of cannister up to 400 yards.

They were also used like ball from 400 to 800 yards, but only against formed troops. Here they specifically used ricochet with the barrel at 6, 10 or 15. At the end of the ricochet the shell was supposed to explode thus causing more casualties. Thus the number of casualties caused should be slightly less than those of a ball (see MW 94 of 1991, plus those shown in table 2).

(Bearing in mind the shell's thin casing it seems to me that on anything like hard ground the case would split and thus one would get neither ricochet nor shell burst --Author).

The main use of howitzers, and therefore of shell, was against stationary targets because of the previously mentioned inaccuracy of the fuse. French shells in particular were badly fused. Shells were therefore used for enfilade and ricochet; to fire buildings, to reach troops behind hills or parapets and in entrenchments; to bound their shells along lines or cavalry; to breach mud walls by exploding their shells in them, and as a primitive sort of air burst against troops in woods. Thus they were particularly useful in sieges firing over the walls as an incendiary device and along the walls against the troops and emplaced artillery, and so were used for anti-morale devices before summoning a garrison to surrender. Because in such cases the target was not moving the fuses could be better adjusted to the range.

As an example of their use against buildings one has only to take the case of Hougoumont during the battle of Waterloo where Napoleon, after his brother Jerome failed to take the farm by assault, ordered howitzers to open fire on the barn and buildings. These were very soon in flames.

Howitzers were also very effective firing cannister, but only at very close range, say 200 yards, because of the wide spread of shot much of the shot projected went in front of, or over, the target. Charles Grant estimated that over 200 yards only about a sixth of the bullets would be effective when fired from a howitzer, but de Toussard states they were used effectively up to 400 yards.

First Shot at Waterloo

The first shot from the Allied side during the battle of Waterloo was a shell which killed three men in the 1 st Light Infantry Regiment of Prince Jerome's Division, and was fired by Major Bull's Troop Royal Horse Artillery (RHA) 3. The troop which consisted of six heavy 5 1/2 inch howitzers also fired over the heads of friendly troops from the height on which it was stationed and took about ten minutes of rapid fire to clear the attacking French brigade out of Hougoumont Wood. The feat was very unusual as the nearer part of the wood was held by friendly forces.

In 1808 during the British invasion of Martinique General Maitland's Division were attacking Fort Desaix where the French had siege and field guns, howitzers and mortars. Five men were killed by a shell and several wounded. (Because of the mix of missiles being fired it is not possible to make any assumption about the calibre or type of weapon responsible, but it can be seen that against troops in close order shell could be very effective - Author).

During the same invasion Browne tells that during a brisk exchange of fire many French shells fell into the British siege lines without doing much execution. They did, however, have some success against a picquet of the 4th West India Regiment where one shell killed 4 men.

In the Peninsular during General Hill's raid on Almaraz in April 1812 the 71 st Highland Light Infantry were advancing on the forts when they were sighted by the French. A couple of shells were fired but no hits were scored. This accords with their normally being used against a stationary target.

Captain (afterwards General) Cavalie Mercer of the Royal Horse Artillery, when covering the withdrawal of the British cavalry from Quatre Bras on 17 June 1815, reported that as several French squadrons approached his G battery of six 9 pounder field guns he gave them a general discharge, then quickly limbered up to retire. Some French horse artillery opened up on the battery before they could complete the manoeuvre, but without much effect. The only person hit was the servant of a Major Whinyates who was wounded in the leg by a shell splinter. (Note he does not tell us anything about the range or the number of rounds fired - Author).

At San Sabastian in 1813 a dozen sailors begged to be allowed to help in the besieging batteries. A shell landed in the battery and one of them attempted to throw it out but it burst in his hand and killed or wounded half of them.

To look at the theoretical performance of the shell there are examples from Prussian records.

Prussian Tests

Lieutenant-General Von Scharnhorst, of the Prussian Artillery, who was mortally wounded at Luetzen in 1813, experimented in Glatz in 1810 to find out the effect on infantry in the open, what effect would occur in masses of earth and in blockhouses, and a shell's general effect in penetration and from it's explosion and splinters.

To judge the effect of splinters he had two concentric circular screens, 6 feet in height, placed at a distance of 20 feet and 40 feet respectively from the centre of the proposed explosions. The effect of 7 and 10 pounder howitzer shells were about equal (the diameter of these shells being 5.66 and 6.55 inches). The explosions threw shell splinters out to 300 paces (250 yards) with ordinary charges, and 460 paces (383 yards) with stronger charges. The ordinary charge of a 7 pounder was 12 ounces of powder and that of a 10 pounder was 1 pound.

The 50 pounder howitzer shell (ten inches in diameter) gave no greater advantage in regard to the number of shell splinters, but it did blow down the screens.

For those of a more technical mind let me quote General Scharnhorst: "The danger of being hit by splinters decreases as the squares of the distance of the station from the shell increase; and is, therefore, in inverse proportion to the squares of the distances of the station from the shell."

He then experimented in a blockhouse 33 1/2 feet long 19 3/4 feet wide and 7 feet high. First he burst a 10 pounder shell with a 1 pound 51/2 ounce charge. This split into 18 or 19 pieces of which he found 16. The blockhouse was very severely shaken and the smoke as they opened the door was very dense and did not disperse through the door and loop-hoes for more than six minutes. The splinters had penetrated deep into the wall and roof of the blockhouse, but had not severely injured them.

He then exploded a 7 pound shell with a 1 pound charge. This burst into 24 pieces of which they found 22. The effect on the blockhouse was not unnaturally less than that of the 10 pounder.

Finally he experimented with what would happen firing into the side of the blockhouse. Firing at 425 paces (354 yards) a 7 pounder, loaded with 3/4 pounds of powder and with a charge of 1 3/4 pounds, penetrated 1 1/2 feet and had nearly broken the beam struck, as well as that laying on its top, and had driven splinters out of the joints. (N.B. The range at which these shells were fired does not accord with the minimum distance given by Wise - perhaps one can assume the shell went off seconds after hitting the target - Author).

A 10 pounder howitzer shell with a charge of 2 1/2 pounds and loaded with 1 pound 4 ounces penetrated 2 feet 8 inches into the wall, dashed to pieces the beam struck, and raised the beams above that smashed. The beams of the blockhouse consisted of the rough trunks of tress which Straith assumed were of fir, which he comments were easier to penetrate than the oak of which ships sides were made.

The 7 (see sketch 3) and 10 pounder Prussian howitzer corresponded respectively and approximately to British 24 and 32 pounder pieces, and to the Gribauval 6 inch howitzer.

Table 1 below gives an idea of some of the different howitzer equipments in use by the main nations involved in the Napoleonic Wars. Howitzers were normally defined as pieces less than twelve calibres in length of the barrel. In general, however, they were between four and six calibres in length.

*The French captured many of these, and found them light and effective. They therefore incorporated them into their System Year Xl (1803) and replaced the heavier 6 inch howitzers in their line and horse artillery 6 pounder batteries. The heavy 6 inch howitzer barrel weighed 701 lbs as opposed to the Austrian Cavalry's 617 lbs.

Country	Piece	Shell caliber and weight	Range in Yards
			Maximum	Effective	Cannister
Austria	7 pdr	6.02 in. 16.5 lbs	1350	700	500
Britain	16 pdr Hy	5.5 in. 15.5 lbs	1700	700	500
France	24 pdr	6.0 in. 23.0 lbs	1200	700	500-600
Prussia	10 pdr	6.55 in.	1950	700	550
Russian	-	10 pdr	1750	700	500-600
Unicorn	-	20 pdr	2000	1100	500-600

Chandler gives the rate of fire of a howitzer as one round per minute for a shell, and a shell burst danger zone as 40 metres in diameter. A French source (Rouquerol) also states 40 metres as the danger zone. Cannister would obviously be faster. In year XIII (1803/4) of the Revolutionary Calendar, Napoleon introduced the heavier 24 pounder howitzer (5.5 inch) into his armies, but never apparently completely replaced the 6 inch model. These latter, as with the guns of the Gribeauval System (4.8 and 12 pdrs), were most likely to be met with in the Peninsular.

There is a clash of facts about the weight of shot fired by the British heavy howitzer, and indeed, even whether the barrel was brass or iron. In the Dickson Manuscripts the weight of the shell is given as 24 pounds and the barrels are stated to be iron by Dickson's superior in England - Major General J. McLeod RA.

Hughes and others give both the heavy and light types as having brass barrels and 16 pound shells. Indeed Hughes says the iron type did not come into use until 1820. (From the amount of detail given by Dickson, and the fact he was a contemporary serving artillery officer of the period, highly thought of by Wellington, I favour his statements - Author).

The Germans do not name their shells from the diameter of the bore, but from the weight of a stone ball the same diameter as the shell. Thus the shell for a 7 pounder howitzer in fact weighs 15 pounds, the 30 lbs howitzer shell weighs 60 pounds and is over 8 inches in diameter. It is probable the Austrians adopted the same practice.

War Game Casualties

Now to suggestions of hits/casualties in a war game. I have found this a most difficult task, and now why many commercial sets of rules duck the subject completely, or wrap shell casualties into the total effect of the gun/battery fire. In previous articles I had more examples and better theory on which to make assumptions. The table below is the result of these considerations. I would be interested in hearing from anyone who has something factual to add. You will note I have not attempted to differentiate between howitzer calibres.

Shellburst
Target Type	Frontal Fire		Enfilade Fire
	% chance of hit	No. of casualties	% chance of hit	No. of casualties
Open Order Infantry & Cavalry	3	1	3	1
Entrenchments; Individual Guns & Wagons; or Fortress Walls	3	6	9	6
A Normal Wall	17	1	3	1
Infantry & Cavalry in Line	17	2	28	4
Artillery deployed	17	1	28	1
Infantry & Cavalry in Column; or Infantry in Square	28	4	28	4
Limbered artillery & Wagons	28	1	58	2
A Building or Bridge	28	5	58	3
A Town or Wood	71	1	71	1

The reason for the unusual percentages is to enable you to use two normal dice, if you do not wish to use percentage dice. Throwing two normal dice then a score of 2 gives a 3% chance; 2 to 3 a 9% chance; 2 to 4 gives 17%; 2 to 5 is 28%; 2 to 7 58%; and 2 to 8 is 71%.

In addition to the above one should take into account the possibility of buildings and towns taking fire, of blowing a hole in a wall as opposed to the shell going through it and exploding inside, and hitting troops behind hills. In the latter case the artillery must have their own observer who can see both the battery firing and the target. In each case I recommend using the next lower percentage chance down. Doing this only one dice throw is necessary to obtain both the hit/casualty chance and that of a fire/hole/troops behind hills. Carry out a similar arithmetical exercise if troops are moving as the table above is based on stationary targets.

The above table represents one minute's fire by one howitzer. To calculate greater periods of fire do NOT increase the percentage chance, but do multiply the casualty figures by the number of minutes. Remember too that the casualties are in men, or in the case of cavalry, men and horses, NOT in figures.

Together with my previous articles there is now enough data on field artillery to cover every army except the British. My next effort will cover spherical case - "shrapnel" thus completing all types of field projectile.