Horse Equipment, Material Culture

Yule’s Modified Field Artillery Harness

United States Field Artillery Harness

Published in the Journal of United State Artillery, 1906, Lt. Yule was attempting to address known problems with the harness then in use. Even in 1906, it appears from reaction to this article that extensive testing and consideration was on-going, testing that would culminate in the results of the 1909 artillery board’s harness design that did away with leather traces and collars entirely. Lt. Yule’s efforts, although notable, were not to yield much fruit. It is interesting to see the equipment in actual use in 1905/6 in the accompanying plates.

MEMBERS of the service are glad to see that due to the rapid advance of the United States field artillery material, we now have a field piece that is equal, if not superior to, those of the European armies. In this advancement there appears to have been an oversight in the matter of improving the locomotive part, i. e. keeping the harness abreast of the latest ideas in equine equipment, in regard to eliminating a useless waste of energy. The following is a report I submitted and some correspondence on my investigations of, and experiments with, field artillery harness.

I have been on duty with the 9th Battery Field Artillery from December 28th, 1902, up to the present time. During that period I have never been quite satisfied with the arrangement of the line of draft (traces from the singletree to the lead-horses’ collar) of the present artillery harness. It is not straight. When all six horses are pulling their proportionate amount of load, there is an unnecessary downward pull on the wheelhorses’ necks. This, in addition to the weight of the collar, pole, and pole-yoke, puts at times an enormous load on the necks of the wheelers. During the two years that I was stationed at Pasay Garrison, Manila, P. I., with the 9th Battery F. A., I seldom noticed a swing or lead-horse with a sore neck. All the wheelhorses were so affected or bore scars on their abused necks. Upon my arrival at the Presidio of San Francisco, Cal., I looked over the horses that the 9th battery took charge of (the old 5th battery animals) and found that almost all of the wheel-horses bore marks of bad necks.

During the last six months’ stay of the battery in the Philippines, by order, it took two fifteen-mile practice marches, __ each month. While on these marches, I paid particular attention to the wheel-horses while going up grades, over soft ground, and crossing bridges. I noticed that if the wheelers pulled their share, they raised the line of draft several inches above the straight line drawn from the singletree to the collar of the lead-horse.

The constant tendency of the wheelers was to lag a little so that the trace-tugs on their collars were almost perpendicular. This would lessen the downward pull on the collar. The wheel-horses are the heaviest of the six and very naturally the largest and tallest. This fact causes the line of draft to be drawn still further out of the straight line that it should occupy; or, in other words, the true line of draft. It frequently happens that when the wheel-horse lags back until he is not pulling a pound and the trace-tug is perpendicular, the line of draft is still too high at this point and there is a downward pull on the wheelers’ necks.

I took the average of several measurements of the height of the trace-plate of the wheel-horses’ collar of the 9th battery F. A. at Pasay Garrison, Manila, P. I. and found that it was 16.5 inches higher than the singletree. The trace-tug is 9.75 inches long; length of wheel-trace and rear connections 78.5 inches; distance from singletree to trace-plate 88.25 inches. Therefore, the height of the point where the swing-trace joins to the wheel-trace is :Ä 85.25 : 16.5 : 78.5 : x = 14.67 inches higher than the singletree.

At this point, the height of the true line of draft above the singletree is as follows :
Distance from singletree to trace-plate of lead-horse 337.75 inches. Height of trace-plate of lead-horse above singletree 16.5 inches. Length of wheel-trace and rear connections 78.5 inches. 337.75 : 16.5 :: 78.5 : x = 3.83 inches.

yule1.gif (17099 bytes)

Therefore, the line of draft is pulled out of its true line, 14.67 – 3.83 = 10.84 inches. This is assuming that the traceplate of the collar of the lead-horse is the same height as that of the wheeler. Usually, it is lower, thus making the error in draft still greater. I took the following average measurements of several sections of the 9th battery F. A. at Pasay Garrison, Manila, P. I., from the harness and horses in use at the time. All in their regular positions ___.

These do not conform exactly with the Ordnance Department measurements on account of stretched leather traces, pole not level, and doubling of trace chains. I assumed the weight of the carriage to be 4500 lbs. (conservative estimate for piece, limber, ammunition, and three men on limber) and assumed a 1 in 10 grade.

From the above, I found that the best authorities on road traction give 200 lbs. per ton as the maximum pull on ordinary earth roads. On grades, the additional traction is equal to 20 lbs. per ton, per per cent of grade, making 200 lbs. per ton on the assumed 1 in 10 grade, altogether 400 lbs. per ton, or a total pull of 900 lbs.

A horse is supposed to exert a pull equal to his weight, with ease, for ten hours a day. This pull then of 900 lbs. gives 150 lbs. per horse, which would be, theoretically, the work of a 1500 lb. horse. The force pulling down on the collar of the wheel-horse then would be the vertical component of the 300 lb. pull of the horses ahead of him. On the assumed 1 in 10 bridge approach, just as the wheelhorse’s collar reached the change from 1 in 10 to level grade this vertical component amounts to about 57 lbs. On a straight 1 in 10 grade it is about 36 lbs. These results are, of course, based upon data obtained from experiments, but are sufficiently exact for all practical purposes.

In starting the load and on rough roads this downward pull would be doubled or even trebled, while on the level it would not be over 15 or 20 lbs. When a battery passes over deep sand going up hill, with all of the horses exerting their full strength, this downward pull becomes so great that the wheelers are simply dragged to earth. I have seen them give away under the strain by crouching or kneeling; or, in other words, progressing by a stumbling gait. In addition to sore necks, the wheel-horses become broken down in front by carrying this load on the fore legs.

From a perusal of the text-books prescribed by the War Department for the use and guidance of mounted troops, I find the following – “HORSES, SADDLES, AND BRIDLES”,CARTER

Page 208. “The serious disadvantages of heavy and cumbersome packs is fully recognized by the authorities in every army, and the problem of devising means whereby the weight may be reduced without decreasing the efficiency of the trooper (applicable as well to artillery soldiers and horses*) is one always open for consideration and experiment.”

Page 111. “While speed is gained for short distances by adjusting the load forward, it will not do for service because of the rapid breaking down which occurs in the fore legs.”

Page 114. “One reason exists for throwing the balance of the horse somewhat to the rear, which alone made it expedient to do so. This is the necessity for preventing the fore feet and legs from becoming prematurely ruined. It is not a matter of theory merely, but a well known result of actual experience, that horses carrying weights upon their backs become broken down in front, as a rule, long before they suffer any deterioration of the hind legs. The date of breaking down is much hastened by saddling far forward (in case of artillery downward pull on collar also*) over the withers, and by an improper use of the stirrups.”

Page 160-1. “Taking a pound out of the amount carried in the horse’s mouth may not seem to be a very important matter at first glance, but when it is considered that a first-class handicapper may, by adding or taking off a pound here or there, entirely upset all calculations as to the result of a race between animals of equal form, it must been seen that a pound more or less, particularly at the end of his neck, makes a great difference to the horse.”

Page 221. “Weight of cavalry kit. The weight of the average kit and equipments complete is about ninety pounds.”
* The writer.

The entire trend of the above extracts is to get weight off of the fore legs of the horse, a few ounces in the bit of the horse being of great moment. Yet, all of our artillery wheel-horses are carrying on their fore legs, in addition to what a cavalry horse does, a fifteen pound collar, pole-yoke, strain from pole, and a downward pull from the lead and swing-horses of 15 to 20 lbs. on the level, to 35 to 57 lbs. on a 1 in 10 grade; and finally, on soft ground up grade an enormous weight which aggregates hundreds of lbs. There is small wonder that a large percentage of wheelhorses have sore necks and are broken down in front.


Page 157. “Harness weights: wheel, near, 71 lbs. 15 ozs. Wheel, off, 69 lbs. 2 ozs.”

Page 173-74. “The loin-straps should be adjusted so that the wheel-traces, when in draft, will be straight and without downward pull on the loops that support them.”
“The loin-straps of swing and lead-horses should raise the traces about six inches above the stifle-joint when in draft. In this position the line of the traces from front to rear will be straight, and the loops of the loin-straps will support the traces without drawing them up.

The traces should be adjusted so that the line of traction will be unbroken from the singletree to the collar of the leaders, and this rule will regulate, in some measure, the length of the loin-straps, the matching of the horses, the arrangement of the pairs as wheel, swing, and lead pair; this should be such as to make the waste of force as small as possible.”

The above description by Dyer seems to overlook the fact that the wheel-horses are bound, by the necessity of the case, to be the largest and their trace-plates the highest. Also, that the singletree is the lowest point of the line of draft. Thus, we have the lowest and highest points of the line of draft very close together, with the necks of the wheelers as a sort of fulcrum over which the lead and swing-horses pull. The average distance of the highest point above the true line of draft, found from my several measurements of different hitched teams, as mentioned before, is 10.84 inches. This error in the line of draft causes an enormous loss of energy. The weight of the harness of the wheelers is almost as much as the entire kit carried by a cavalry horse. From the above discussion, it is evident that the artillery horse is required to carry much more weight than the cavalry horse and, this in addition to his first duty of drawing the carriage.

In movement of troops the first and all important thing is to cover the maximum distance with the minimum loss of energy. To accomplish this end, I have devised a line of draft and have had it in successful use in the 9th battery F. A., since June 3d, 1905, which eliminates the downward pull on the wheelers’ necks. It is simple, requires but very little material than the present harness, and, does not change any parts

The battery has operated over the steep and rugged hills of the Presidio of San Francisco; gone through deep sand; taken 15-mile practice marches; and the device has worked satisfactorily. I gave it an exhaustive test in counter-marching, reversing, etc., at a trot and gallop and found it to work perfectly. The description of the device is as follows: (Particulars can be easily seen in the accompanying photographs).

Take two regulation wheel-traces (leather or steel cable) join them together at the rear ends by a cockeye passed through the two D rings in such a manner that the traces will be close together but not rub. This cockeye is plainly seen in photograph 2. It is made of 1-in. best quality Norway iron, 11 in., long and bent as shown in the photograph. In case of leather traces put a strip of leather around rear end of each to prevent wearing on each other, as seen in photograph 2.

The trace that is attached to the wheel-horse to remain intact. Lengthen the other trace eight and one-half inches, by inserting more links in the front chain; (in case of adoption, lengthen the body of the trace instead). This makes it long enough to bring the toggle directly underneath the trace-plate of the wheeler’s collar.

Take a breast strap, or similar strap, leave one breast strap hook on it; engage the hook in the breast-strap ring of the trace-plate (there is room for it); put a double eye-loop, similar to the one on the trace-tug, on the strap and buckle it up; engage the toggle of the longest trace in this eye-loop. This strap should be adjustable in order that the lower trace can be brought into the true line of draft for different height horses. It should be about thirty or thirty-two inches long.

Make a double loop on the loin-strap. The lower loop about six or eight inches long. Eight inches will give an adjustment of sixteen inches separation at the front end of the traces. Then one adjustment of the loin-strap would do for both traces.

From the above, it is readily seen that to put the device in in use, it is not necessary to manufacture any different material nor to change any now in use except to lengthen one trace, manufacture a double cockeye which I have made, and put a double loop on the loin-strap. In case of adoption, there should be another ring put on the trace-plate.

In order to give the device a practical test, it is not necessary to lengthen the lower trace. Simply arrange the harness as shown in photograph 2. In Plate II both traces are the same length, (regulation traces) the only difference being that the supporting strap slants backward instead of being vertical. This is not so good as the other, however, as it strikes the horse’s fore leg too much. The large dray and transfer companies have solved this problem of getting the weight off of the horses necks by having the pole held up either by a spring or patent hounds. The doubletrees for the lead-horses are put on the end of the pole.

The wheel-horses’ breechings are independent of their collars. This gives the wheelers complete freedom of movement and absolutely no weight but that of the collar, on the necks. This arrangement is impracticable in artillery. Therefore, the next best thing is the double-trace.

A careful study of the English and French breechings, convinces me that the one now in use in the U. S. Artillery, is the better. The advantages of this double-trace, are:

I. There is no downward pull on the wheeler’s necks from the leaders.
II. It allows the wheeler to exert more of an upward pull on the carriage when it drops into a rut, low place, or sinks into soft ground, i.e. the same lifting force that is given in working on the rear of a wheel by the cannoneers No. 1 and 2 in “By hand to the front.”
III. Elimination of the jerk on the wheelers’ necks in going down grade or across ditches, if the lead and swing-horses go into draft as they are bound to, and often do after they have crossed a ditch, recovered themselves and the wheelers are still in the ditch and generally not in draft. By the lead and swing-horses having an independent line of draft they come into it, keep the carriage moving at a critical moment, after they have crossed and fully recovered themselves. The wheelers are not interfered with as they are crossing the bad place, have freedom of movement in coming up the bank and come into draft gradually after having recovered themselves from holding the carriage back and getting through the ditch.
IV. Elimination of see-saw motion on the wheelers’ necks from the lead and swing-horses not keeping in step. This is considerable when all are in draft, as they should be.
V. The wheelers have complete freedom of motion at all times, especially when going at a gallop.
VI. The wheelers are not held down with a load on their necks from the pull of the horses in front of them.
VII. It lowers the traces of the swing-horses so that their collars do not slip upward, when in draft, and choke them.
VIII. Everything presented in this paper, and many other things that could be mentioned by a further study of the subject.

I am fully convinced that this device in an improvement on the present field artillery harness, and therefore request that it be given a trial in the two Provisional Regiments of Field Artillery recently organized.

I took a copy of the above report, a 3.2-inch field piece, and the harness to Headquarters Pacific Division, San Francisco, Cal., for the officers to inspect. A discussion arose as to the weight I had assumed for the load drawn by a six-horse team of artillery horses. In defense of my assumption, I wrote the following letter:

Presidio of San Francisco,
San Francisco, Cal., Aug. 11, 1905.
Col. Stephen P. Jocelyn,
San Francisco, Cal.,

I beg leave to write to you, informally, about the harness that I had down at the Division Headquarters last week.

During the reading of my report there, a discussion arose about the weight of the load that the six horses were required to pull. I had assumed that the weight was 4500 pounds. The officers were inclined to think that that was too much.

The next day, after weekly inspection, I weighed the gun that I had downtown for the officers to see. I unhitched the horses, and mounted three men on the limber chest. I took the three men that happened to be with the gun. The gun was weighed on the scales on which the forage of the Post is weighed. Then I figured the weight of ammunition and powder that would be in the limber in case that the battery was equipped for service. Following are the weights:

3822 lbs. weight of gun and carriage with three men mounted.
567 lbs. weight of 42 shell and shrapnel, each 13.5 lbs.
34.5 lbs. weight of three canister that are on the trail of the piece, each 11 lbs. 8 ozs.
43.5 lbs. weight of 45 rounds of powder, each 15.5 ozs.
4467.0 lbs. Total

That is the weight of each of the guns in the two batteries that I have been with during the last three years. Now this weight is still greater at times. When a battery is prepared for action, there are two more men mounted on the axle seats, which increases the weight about three hundred pounds. In wet weather when the gun and carriage gets watersoaked they will become still heavier. So you can see my assumption of 4500 lbs. was, as I stated, conservative.

The last two days the field artillery battalion has gone out to drill here with the two men mounted on the axle seats. To-day we went up the beach toward Fort Point (through deep sand) then up a steep hill, a distance of fully a mile. Part of the distance up the hill was covered at a trot. Of course, at present, we have no ammunition in the limbers. But if it had been service conditions all of the pieces would have weighed nearly 4800 lbs. each.

When on marches, it is customary for the men to be allowed to ride on the axle seats to smoke. This is done in order to safe-guard the ammunition in the chests. So you see a good share of the time on the march there is this weight of almost 4800 lbs. to be drawn by the piece horses.

The above is not theory, but facts, as they exist. I have been unable to obtain weight of the new equipment but suspect that it has been considerably lightened, as I remember of having read of the wheels being lightened.

Hoping that this will sustain me in the article, I am,
Very respectfully,
1st Lieutenant, Artillery Corps.

On September 4th, 1905, the First Battalion Field Artillery left Presidio of San Francisco, Cal., on its annual practice march. I had command of the 9th battery, and previous to starting had the wheel-horses of the four pieces equipped with the double-traces. I changed neither any horses nor drivers to favor the device but let it stand on its own merits.

Upon arrival in camp, after a 225-m march, I checked up the sore necks of the wheelers and found:
Wheelers of pieces: one sore, two scalds.
Wheelers of caissons, battery and forage wagons three sore, nine scalds, and two chafed.
That is, there were three-eights of the double-trace and seven-tenths of the others with worsted necks.

Coming back the same distance, I had:
Wheelers of pieces: one small boil (not deep seated sore, but a pimple), one scald, and one bruised.
Wheelers of caissons, battery and forage’ wagons: one small boil, (pimple) six scalded, and two bruised
That is, there were three-eighths of the double-trace and nine-twentieths of the others worsted.
The wheelers of the pieces were at a disadvantage during the march on account of there being no brakes on the piece carriages.

It will be noticed that there was a smaller percentage of sore necks in the twenty caisson wheelers returning than going out. The results of the double-trace were so satisfactory going out that some of the caisson drivers asked permission to have double-traces. I was naturally pleased at this, and had all of my spare traces utilized, and fitted up five caisson wheelers with the double-trace, on the return trip.

Some of the wheel-horses of the caissons were occasionally changed with the swing-horses. Four of the piece wheelers worked in their places the entire six weeks that we were on the march and in the field; two nearly all the time, and the remaining two the same as the caisson wheelers were. The results obtained prompt me to publish them. I received the following letter, on October 20, upon return from the march:

War Department,
The Military Secretary’s Office.
Washington, September 14, 1905.
1st Lieutenant, Edgar H. Yule, Artillery Corps, * Presidio of San Francisco, Cal.
(Through Headquarters, Department of California.)

In response to your letter of July 15th, last, in which you submit a description, with photographs, of modification of the present harness for wheel-horses of field batteries devised by you, the the honor to inform you that the professional zeal displayed by you in devising a modification of the present harness for wheel-horses of field artillery is commended by the Acting Chief `of Staff, but that another device practically accomplishing the same purpose, without so great an addition to the weight of the harness is now under trial and is preferable to the scheme proposed by you.
Very respectfully,
Military Secretary. (S.W.D.)

In reference to the above letter, in regard to the device not being desirable on account of its weight, I beg leave to say: As shown in the photographs it is too heavy. It was gotten up with the material at hand. But the theory of the double-trace is feasible and an improvement.

The supporting strap and metal connections in front can be reduced more than two-thirds in size and weight and still be strong enough. It is subjected to no stress except an occasional jerk and in coming on a level from an incline, viz: going on to a bridge. A bridle rein strap is strong enough.

The trace proper can be lightened. The upper one now having but one horse working on it can be reduced two-thirds in weight. The lower one having but two horses working on it can be reduced one-third. The new wire-rope wheel trace weighs 2 lbs. 14 oz. and the former leather trace 3 lbs. 12 oz. Put on two wire-rope traces with the above mentioned percentage of reductions and you have a double-trace weighing 14 oz. less than the old leather one and the same weight as the present single wire-rope trace. Therefore, considering the vast advantage the wheelers have, as shown in the eight “advantages” given above, I am quite sure that the device is superior to any other that can be put in use, or to the present harness.

Another added advantage to the eight above mentioned is, that there is no need of the detachable trace-chains, thus reducing the weight of each trace 1 lb. 2 oz.
I have never been able to see the utility of these chains for the present harness, except to be constantly getting lost and delaying getting hitched in, and there is certainly no need for them with the double-trace. (Some battery commanders remove the rear trace-chains of the wheel-traces. This is the case in the accompanying photographs.)yuleplate1yuleplate2yuleplate3

They are intended to adjust the length of the traces for different sized horses, viz :Drill Regulations Field Artillery, 1905. (Provisional page 96.) “The length of the traces must depend in a great measure on the size of the horse and his stride. For the wheel team, the rule is to allow about fourteen inches from the singletree to hindquarters, and for swing and lead teams, about one yard from head to point of buttocks when in draft. The traces should be adjusted so that the line of traction will be straight from the singletree to the collars of the leaders,” (last sentence impossible with present single trace).

A moment’s thought will make it clear that if the wheeltraces are to be adjusted for different sized horses, the pole will also have to be adjusted, in length, as it is far more important that the collars be in such a position that the pole-yoke will be held at the proper angle in relation to the pole-yoke stop, than that the horse’s hindquarters be a trifle of a few inches farther from or nearer to the singletree.

The only adjustments for the longitudinal stresses made necessary by different sized horses are the breeching and a slight change of the breast strap to keep the pole at the proper height. And in the case of small horses care must be taken or the latter will be gotten too short and bring all the hold-back on the horse’s neck instead of the breeching. These chains become worn so that they become unhooked very easily. Men resort to all sorts of schemes to make them secure, even to tying knots in them.

On account of these chains (in the Philippines, where the leather traces stretched unduly) it was necessary to strap the pole-yoke to the pole to prevent the latter from dropping but of the former. Finally, the chains were discarded on the wheel horses. With the new wire-rope trace, where the liability to stretch is reduced to a minimum, there is no necessity for them to be adjusted.

The collar should always bear the same relation to the end of the pole. The difference caused by the height o different sized horses is so small that it can be easily corrected by the breast straps simply to keep the pole at the proper height In photograph No. 3 a line AB is drawn from the end of the doubletree A, to the trace-plate of the collar B ; another, BC, from the trace-plate B, slanting down and forward to the pole-yoke stop, C, on the end of the pole; and one from C to A, making a triangle. It will be seen that angle B is a little more than a right, angle. Extend B horizontally forward to B’. B’ is a right angle. B would be a little beyond B’ at B” if the horse had been standing well in draft at the instant the photograph was taken. Angle AB’ `C should equal angle ACB”.

With these two angles equal you have an isosceles triangle in which the trace and connections are one side, i. e. AB” The other side is the line AC; i. e. in reality, the length of the pole from the doubletree bolt to the neck-yoke stop. Now this isosceles triangle is lying on the side AC, which is practically horizontal and in most cases C is slightly higher than A. Therefore, the angle B” will always be farther to the rear than angle C. That is to say: No matter how large or small the horse the isosceles triangle simply has a longer or shorter base, B’C, caused by adjusting the breast strap, and the collar will approach or recede from the pole-yoke stop and the angles B” and C remain equal. The collar never gets forward of the pole-yoke stop nor too far to the rear of it.

Thus far I have used only my own measurements and a graphic representation. I will now take some Ordnance Department measurements of the new field artillery material and show that this isosceles triangle now exists; but all of its advantages are lost by the provisions of the drill regulations which permit the driver to adjust the length of his trace with the new trace-chain when he imagines that something is wrong, instead of using the proper remedy; i. e. the breast strap and side straps of the breeching.

The new pole from the doubletree bolt to the pole-yoke stop is 102.687 inches. (Blue print 3-inch field limber, model 1902, R.I.A.)
The new steel wire trace with connections from end of doubletree to trace-plate of collar is:
8.25 inches doubletree to end of whippletree hook.
11. 625 inches rear chain.
9.75 inches Mogul spring.
58. inches trace.
5. inches front chain.
9.75 inches tug.

102.375 inches equals AB” (blue print, artillery harness traces R.I.A. 1904.) The pole equals AC. The trace and connections equals AB” of the established triangle, and is only .312 of an inch shorter than AC. I do not know whether the Ordnance Department intended this triangle to be or not. Nevertheless, it does exist when the harness is properly adjusted. Therefore, the rear chain should be discarded and the trace lengthened. With the non-stretching trace, the length needs no more adjusting than the piston rod of an engine for a varying steam pressure.

With the double-trace the wheel horse is independent of the others. On all kinds of roads he performs his functions without hinderance; and at the end of a long march his driver can feel that the animal’s great strength has been utilized to the best advantage, and that with a minimum loss of energy.